quoccuongle/Grounded-SAM-2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

init commit of samurai

Browse Source
This commit is contained in:
Cheng-Yen Yang
2024-11-19 22:12:54 -08:00
parent f65f4ba181
commit c17e4cecc0
679 changed files with 123982 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

85
sam2/.clang-format Normal file
View File

@@ -0,0 +1,85 @@
AccessModifierOffset: -1
AlignAfterOpenBracket: AlwaysBreak
AlignConsecutiveAssignments: false
AlignConsecutiveDeclarations: false
AlignEscapedNewlinesLeft: true
AlignOperands: false
AlignTrailingComments: false
AllowAllParametersOfDeclarationOnNextLine: false
AllowShortBlocksOnASingleLine: false
AllowShortCaseLabelsOnASingleLine: false
AllowShortFunctionsOnASingleLine: Empty
AllowShortIfStatementsOnASingleLine: false
AllowShortLoopsOnASingleLine: false
AlwaysBreakAfterReturnType: None
AlwaysBreakBeforeMultilineStrings: true
AlwaysBreakTemplateDeclarations: true
BinPackArguments: false
BinPackParameters: false
BraceWrapping:
AfterClass: false
AfterControlStatement: false
AfterEnum: false
AfterFunction: false
AfterNamespace: false
AfterObjCDeclaration: false
AfterStruct: false
AfterUnion: false
BeforeCatch: false
BeforeElse: false
IndentBraces: false
BreakBeforeBinaryOperators: None
BreakBeforeBraces: Attach
BreakBeforeTernaryOperators: true
BreakConstructorInitializersBeforeComma: false
BreakAfterJavaFieldAnnotations: false
BreakStringLiterals: false
ColumnLimit: 80
CommentPragmas: '^ IWYU pragma:'
ConstructorInitializerAllOnOneLineOrOnePerLine: true
ConstructorInitializerIndentWidth: 4
ContinuationIndentWidth: 4
Cpp11BracedListStyle: true
DerivePointerAlignment: false
DisableFormat: false
ForEachMacros: [ FOR_EACH, FOR_EACH_R, FOR_EACH_RANGE, ]
IncludeCategories:
- Regex: '^<.*\.h(pp)?>'
Priority: 1
- Regex: '^<.*'
Priority: 2
- Regex: '.*'
Priority: 3
IndentCaseLabels: true
IndentWidth: 2
IndentWrappedFunctionNames: false
KeepEmptyLinesAtTheStartOfBlocks: false
MacroBlockBegin: ''
MacroBlockEnd: ''
MaxEmptyLinesToKeep: 1
NamespaceIndentation: None
ObjCBlockIndentWidth: 2
ObjCSpaceAfterProperty: false
ObjCSpaceBeforeProtocolList: false
PenaltyBreakBeforeFirstCallParameter: 1
PenaltyBreakComment: 300
PenaltyBreakFirstLessLess: 120
PenaltyBreakString: 1000
PenaltyExcessCharacter: 1000000
PenaltyReturnTypeOnItsOwnLine: 200
PointerAlignment: Left
ReflowComments: true
SortIncludes: true
SpaceAfterCStyleCast: false
SpaceBeforeAssignmentOperators: true
SpaceBeforeParens: ControlStatements
SpaceInEmptyParentheses: false
SpacesBeforeTrailingComments: 1
SpacesInAngles: false
SpacesInContainerLiterals: true
SpacesInCStyleCastParentheses: false
SpacesInParentheses: false
SpacesInSquareBrackets: false
Standard: Cpp11
TabWidth: 8
UseTab: Never

17
sam2/.github/workflows/check_fmt.yml vendored Normal file
View File

@@ -0,0 +1,17 @@
name: SAM2/fmt
on:
pull_request:
branches:
- main
jobs:
ufmt_check:
runs-on: ubuntu-latest
steps:
- name: Check formatting
uses: omnilib/ufmt@action-v1
with:
path: sam2 tools
version: "2.0.0b2"
python-version: "3.10"
black-version: "24.2.0"
usort-version: "1.0.2"

10
sam2/.gitignore vendored Normal file
View File

@@ -0,0 +1,10 @@
.vscode/
.DS_Store
__pycache__/
*-checkpoint.ipynb
.venv
*.egg*
build/*
_C.*
outputs/*
checkpoints/*.pt

1
sam2/.watchmanconfig Normal file
View File

@@ -0,0 +1 @@
{}

80
sam2/CODE_OF_CONDUCT.md Normal file
View File

@@ -0,0 +1,80 @@
# Code of Conduct
## Our Pledge
In the interest of fostering an open and welcoming environment, we as
contributors and maintainers pledge to make participation in our project and
our community a harassment-free experience for everyone, regardless of age, body
size, disability, ethnicity, sex characteristics, gender identity and expression,
level of experience, education, socio-economic status, nationality, personal
appearance, race, religion, or sexual identity and orientation.
## Our Standards
Examples of behavior that contributes to creating a positive environment
include:
* Using welcoming and inclusive language
* Being respectful of differing viewpoints and experiences
* Gracefully accepting constructive criticism
* Focusing on what is best for the community
* Showing empathy towards other community members
Examples of unacceptable behavior by participants include:
* The use of sexualized language or imagery and unwelcome sexual attention or
advances
* Trolling, insulting/derogatory comments, and personal or political attacks
* Public or private harassment
* Publishing others' private information, such as a physical or electronic
address, without explicit permission
* Other conduct which could reasonably be considered inappropriate in a
professional setting
## Our Responsibilities
Project maintainers are responsible for clarifying the standards of acceptable
behavior and are expected to take appropriate and fair corrective action in
response to any instances of unacceptable behavior.
Project maintainers have the right and responsibility to remove, edit, or
reject comments, commits, code, wiki edits, issues, and other contributions
that are not aligned to this Code of Conduct, or to ban temporarily or
permanently any contributor for other behaviors that they deem inappropriate,
threatening, offensive, or harmful.
## Scope
This Code of Conduct applies within all project spaces, and it also applies when
an individual is representing the project or its community in public spaces.
Examples of representing a project or community include using an official
project e-mail address, posting via an official social media account, or acting
as an appointed representative at an online or offline event. Representation of
a project may be further defined and clarified by project maintainers.
This Code of Conduct also applies outside the project spaces when there is a
reasonable belief that an individual's behavior may have a negative impact on
the project or its community.
## Enforcement
Instances of abusive, harassing, or otherwise unacceptable behavior may be
reported by contacting the project team at <opensource-conduct@fb.com>. All
complaints will be reviewed and investigated and will result in a response that
is deemed necessary and appropriate to the circumstances. The project team is
obligated to maintain confidentiality with regard to the reporter of an incident.
Further details of specific enforcement policies may be posted separately.
Project maintainers who do not follow or enforce the Code of Conduct in good
faith may face temporary or permanent repercussions as determined by other
members of the project's leadership.
## Attribution
This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4,
available at https://www.contributor-covenant.org/version/1/4/code-of-conduct.html
[homepage]: https://www.contributor-covenant.org
For answers to common questions about this code of conduct, see
https://www.contributor-covenant.org/faq

31
sam2/CONTRIBUTING.md Normal file
View File

@@ -0,0 +1,31 @@
# Contributing to segment-anything
We want to make contributing to this project as easy and transparent as
possible.
## Pull Requests
We actively welcome your pull requests.
1. Fork the repo and create your branch from `main`.
2. If you've added code that should be tested, add tests.
3. If you've changed APIs, update the documentation.
4. Ensure the test suite passes.
5. Make sure your code lints, using the `ufmt format` command. Linting requires `black==24.2.0`, `usort==1.0.2`, and `ufmt==2.0.0b2`, which can be installed via `pip install -e ".[dev]"`.
6. If you haven't already, complete the Contributor License Agreement ("CLA").
## Contributor License Agreement ("CLA")
In order to accept your pull request, we need you to submit a CLA. You only need
to do this once to work on any of Facebook's open source projects.
Complete your CLA here: <https://code.facebook.com/cla>
## Issues
We use GitHub issues to track public bugs. Please ensure your description is
clear and has sufficient instructions to be able to reproduce the issue.
Facebook has a [bounty program](https://www.facebook.com/whitehat/) for the safe
disclosure of security bugs. In those cases, please go through the process
outlined on that page and do not file a public issue.
## License
By contributing to segment-anything, you agree that your contributions will be licensed
under the LICENSE file in the root directory of this source tree.

189
sam2/INSTALL.md Normal file
View File

@@ -0,0 +1,189 @@
## Installation
### Requirements
- Linux with Python ≥ 3.10, PyTorch ≥ 2.3.1 and [torchvision](https://github.com/pytorch/vision/) that matches the PyTorch installation. Install them together at https://pytorch.org to ensure this.
* Note older versions of Python or PyTorch may also work. However, the versions above are strongly recommended to provide all features such as `torch.compile`.
- [CUDA toolkits](https://developer.nvidia.com/cuda-toolkit-archive) that match the CUDA version for your PyTorch installation. This should typically be CUDA 12.1 if you follow the default installation command.
- If you are installing on Windows, it's strongly recommended to use [Windows Subsystem for Linux (WSL)](https://learn.microsoft.com/en-us/windows/wsl/install) with Ubuntu.
Then, install SAM 2 from the root of this repository via
```bash
pip install -e ".[notebooks]"
```
Note that you may skip building the SAM 2 CUDA extension during installation via environment variable `SAM2_BUILD_CUDA=0`, as follows:
```bash
# skip the SAM 2 CUDA extension
SAM2_BUILD_CUDA=0 pip install -e ".[notebooks]"
```
This would also skip the post-processing step at runtime (removing small holes and sprinkles in the output masks, which requires the CUDA extension), but shouldn't affect the results in most cases.
### Building the SAM 2 CUDA extension
By default, we allow the installation to proceed even if the SAM 2 CUDA extension fails to build. (In this case, the build errors are hidden unless using `-v` for verbose output in `pip install`.)
If you see a message like `Skipping the post-processing step due to the error above` at runtime or `Failed to build the SAM 2 CUDA extension due to the error above` during installation, it indicates that the SAM 2 CUDA extension failed to build in your environment. In this case, **you can still use SAM 2 for both image and video applications**. The post-processing step (removing small holes and sprinkles in the output masks) will be skipped, but this shouldn't affect the results in most cases.
If you would like to enable this post-processing step, you can reinstall SAM 2 on a GPU machine with environment variable `SAM2_BUILD_ALLOW_ERRORS=0` to force building the CUDA extension (and raise errors if it fails to build), as follows
```bash
pip uninstall -y SAM-2 && \
rm -f ./sam2/*.so && \
SAM2_BUILD_ALLOW_ERRORS=0 pip install -v -e ".[notebooks]"
```
Note that PyTorch needs to be installed first before building the SAM 2 CUDA extension. It's also necessary to install [CUDA toolkits](https://developer.nvidia.com/cuda-toolkit-archive) that match the CUDA version for your PyTorch installation. (This should typically be CUDA 12.1 if you follow the default installation command.) After installing the CUDA toolkits, you can check its version via `nvcc --version`.
Please check the section below on common installation issues if the CUDA extension fails to build during installation or load at runtime.
### Common Installation Issues
Click each issue for its solutions:
<details>
<summary>
I got `ImportError: cannot import name '_C' from 'sam2'`
</summary>
<br/>
This is usually because you haven't run the `pip install -e ".[notebooks]"` step above or the installation failed. Please install SAM 2 first, and see the other issues if your installation fails.
In some systems, you may need to run `python setup.py build_ext --inplace` in the SAM 2 repo root as suggested in https://github.com/facebookresearch/sam2/issues/77.
</details>
<details>
<summary>
I got `MissingConfigException: Cannot find primary config 'configs/sam2.1/sam2.1_hiera_l.yaml'`
</summary>
<br/>
This is usually because you haven't run the `pip install -e .` step above, so `sam2` isn't in your Python's `sys.path`. Please run this installation step. In case it still fails after the installation step, you may try manually adding the root of this repo to `PYTHONPATH` via
```bash
export SAM2_REPO_ROOT=/path/to/sam2 # path to this repo
export PYTHONPATH="${SAM2_REPO_ROOT}:${PYTHONPATH}"
```
to manually add `sam2_configs` into your Python's `sys.path`.
</details>
<details>
<summary>
I got `RuntimeError: Error(s) in loading state_dict for SAM2Base` when loading the new SAM 2.1 checkpoints
</summary>
<br/>
This is likely because you have installed a previous version of this repo, which doesn't have the new modules to support the SAM 2.1 checkpoints yet. Please try the following steps:
1. pull the latest code from the `main` branch of this repo
2. run `pip uninstall -y SAM-2` to uninstall any previous installations
3. then install the latest repo again using `pip install -e ".[notebooks]"`
In case the steps above still don't resolve the error, please try running in your Python environment the following
```python
from sam2.modeling import sam2_base
print(sam2_base.__file__)
```
and check whether the content in the printed local path of `sam2/modeling/sam2_base.py` matches the latest one in https://github.com/facebookresearch/sam2/blob/main/sam2/modeling/sam2_base.py (e.g. whether your local file has `no_obj_embed_spatial`) to indentify if you're still using a previous installation.
</details>
<details>
<summary>
My installation failed with `CUDA_HOME environment variable is not set`
</summary>
<br/>
This usually happens because the installation step cannot find the CUDA toolkits (that contain the NVCC compiler) to build a custom CUDA kernel in SAM 2. Please install [CUDA toolkits](https://developer.nvidia.com/cuda-toolkit-archive) or the version that matches the CUDA version for your PyTorch installation. If the error persists after installing CUDA toolkits, you may explicitly specify `CUDA_HOME` via
```
export CUDA_HOME=/usr/local/cuda # change to your CUDA toolkit path
```
and rerun the installation.
Also, you should make sure
```
python -c 'import torch; from torch.utils.cpp_extension import CUDA_HOME; print(torch.cuda.is_available(), CUDA_HOME)'
```
print `(True, a directory with cuda)` to verify that the CUDA toolkits are correctly set up.
If you are still having problems after verifying that the CUDA toolkit is installed and the `CUDA_HOME` environment variable is set properly, you may have to add the `--no-build-isolation` flag to the pip command:
```
pip install --no-build-isolation -e .
```
</details>
<details>
<summary>
I got `undefined symbol: _ZN3c1015SmallVectorBaseIjE8grow_podEPKvmm` (or similar errors)
</summary>
<br/>
This usually happens because you have multiple versions of dependencies (PyTorch or CUDA) in your environment. During installation, the SAM 2 library is compiled against one version library while at run time it links against another version. This might be due to that you have different versions of PyTorch or CUDA installed separately via `pip` or `conda`. You may delete one of the duplicates to only keep a single PyTorch and CUDA version.
In particular, if you have a lower PyTorch version than 2.3.1, it's recommended to upgrade to PyTorch 2.3.1 or higher first. Otherwise, the installation script will try to upgrade to the latest PyTorch using `pip`, which could sometimes lead to duplicated PyTorch installation if you have previously installed another PyTorch version using `conda`.
We have been building SAM 2 against PyTorch 2.3.1 internally. However, a few user comments (e.g. https://github.com/facebookresearch/sam2/issues/22, https://github.com/facebookresearch/sam2/issues/14) suggested that downgrading to PyTorch 2.1.0 might resolve this problem. In case the error persists, you may try changing the restriction from `torch>=2.3.1` to `torch>=2.1.0` in both [`pyproject.toml`](pyproject.toml) and [`setup.py`](setup.py) to allow PyTorch 2.1.0.
</details>
<details>
<summary>
I got `CUDA error: no kernel image is available for execution on the device`
</summary>
<br/>
A possible cause could be that the CUDA kernel is somehow not compiled towards your GPU's CUDA [capability](https://developer.nvidia.com/cuda-gpus). This could happen if the installation is done in an environment different from the runtime (e.g. in a slurm system).
You can try pulling the latest code from the SAM 2 repo and running the following
```
export TORCH_CUDA_ARCH_LIST=9.0 8.0 8.6 8.9 7.0 7.2 7.5 6.0`
```
to manually specify the CUDA capability in the compilation target that matches your GPU.
</details>
<details>
<summary>
I got `RuntimeError: No available kernel. Aborting execution.` (or similar errors)
</summary>
<br/>
This is probably because your machine doesn't have a GPU or a compatible PyTorch version for Flash Attention (see also https://discuss.pytorch.org/t/using-f-scaled-dot-product-attention-gives-the-error-runtimeerror-no-available-kernel-aborting-execution/180900 for a discussion in PyTorch forum). You may be able to resolve this error by replacing the line
```python
OLD_GPU, USE_FLASH_ATTN, MATH_KERNEL_ON = get_sdpa_settings()
```
in [`sam2/modeling/sam/transformer.py`](sam2/modeling/sam/transformer.py) with
```python
OLD_GPU, USE_FLASH_ATTN, MATH_KERNEL_ON = True, True, True
```
to relax the attention kernel setting and use other kernels than Flash Attention.
</details>
<details>
<summary>
I got `Error compiling objects for extension`
</summary>
<br/>
You may see error log of:
> unsupported Microsoft Visual Studio version! Only the versions between 2017 and 2022 (inclusive) are supported! The nvcc flag '-allow-unsupported-compiler' can be used to override this version check; however, using an unsupported host compiler may cause compilation failure or incorrect run time execution. Use at your own risk.
This is probably because your versions of CUDA and Visual Studio are incompatible. (see also https://stackoverflow.com/questions/78515942/cuda-compatibility-with-visual-studio-2022-version-17-10 for a discussion in stackoverflow).<br>
You may be able to fix this by adding the `-allow-unsupported-compiler` argument to `nvcc` after L48 in the [setup.py](https://github.com/facebookresearch/sam2/blob/main/setup.py). <br>
After adding the argument, `get_extension()` will look like this:
```python
def get_extensions():
srcs = ["sam2/csrc/connected_components.cu"]
compile_args = {
"cxx": [],
"nvcc": [
"-DCUDA_HAS_FP16=1",
"-D__CUDA_NO_HALF_OPERATORS__",
"-D__CUDA_NO_HALF_CONVERSIONS__",
"-D__CUDA_NO_HALF2_OPERATORS__",
"-allow-unsupported-compiler" # Add this argument
],
}
ext_modules = [CUDAExtension("sam2._C", srcs, extra_compile_args=compile_args)]
return ext_modules
```
</details>

201
sam2/LICENSE Normal file
View File

@@ -0,0 +1,201 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

29
sam2/LICENSE_cctorch Normal file
View File

@@ -0,0 +1,29 @@
BSD 3-Clause License
Copyright (c) 2020, the respective contributors, as shown by the AUTHORS file.
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

7
sam2/MANIFEST.in Normal file
View File

@@ -0,0 +1,7 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
recursive-include sam2 *.yaml #include all config files

219
sam2/README.md Normal file
View File

@@ -0,0 +1,219 @@
# SAM 2: Segment Anything in Images and Videos
**[AI at Meta, FAIR](https://ai.meta.com/research/)**
[Nikhila Ravi](https://nikhilaravi.com/), [Valentin Gabeur](https://gabeur.github.io/), [Yuan-Ting Hu](https://scholar.google.com/citations?user=E8DVVYQAAAAJ&hl=en), [Ronghang Hu](https://ronghanghu.com/), [Chaitanya Ryali](https://scholar.google.com/citations?user=4LWx24UAAAAJ&hl=en), [Tengyu Ma](https://scholar.google.com/citations?user=VeTSl0wAAAAJ&hl=en), [Haitham Khedr](https://hkhedr.com/), [Roman Rädle](https://scholar.google.de/citations?user=Tpt57v0AAAAJ&hl=en), [Chloe Rolland](https://scholar.google.com/citations?hl=fr&user=n-SnMhoAAAAJ), [Laura Gustafson](https://scholar.google.com/citations?user=c8IpF9gAAAAJ&hl=en), [Eric Mintun](https://ericmintun.github.io/), [Junting Pan](https://junting.github.io/), [Kalyan Vasudev Alwala](https://scholar.google.co.in/citations?user=m34oaWEAAAAJ&hl=en), [Nicolas Carion](https://www.nicolascarion.com/), [Chao-Yuan Wu](https://chaoyuan.org/), [Ross Girshick](https://www.rossgirshick.info/), [Piotr Dollár](https://pdollar.github.io/), [Christoph Feichtenhofer](https://feichtenhofer.github.io/)
[[`Paper`](https://ai.meta.com/research/publications/sam-2-segment-anything-in-images-and-videos/)] [[`Project`](https://ai.meta.com/sam2)] [[`Demo`](https://sam2.metademolab.com/)] [[`Dataset`](https://ai.meta.com/datasets/segment-anything-video)] [[`Blog`](https://ai.meta.com/blog/segment-anything-2)] [[`BibTeX`](#citing-sam-2)]
![SAM 2 architecture](assets/model_diagram.png?raw=true)
**Segment Anything Model 2 (SAM 2)** is a foundation model towards solving promptable visual segmentation in images and videos. We extend SAM to video by considering images as a video with a single frame. The model design is a simple transformer architecture with streaming memory for real-time video processing. We build a model-in-the-loop data engine, which improves model and data via user interaction, to collect [**our SA-V dataset**](https://ai.meta.com/datasets/segment-anything-video), the largest video segmentation dataset to date. SAM 2 trained on our data provides strong performance across a wide range of tasks and visual domains.
![SA-V dataset](assets/sa_v_dataset.jpg?raw=true)
## Latest updates
**09/30/2024 -- SAM 2.1 Developer Suite (new checkpoints, training code, web demo) is released**
- A new suite of improved model checkpoints (denoted as **SAM 2.1**) are released. See [Model Description](#model-description) for details.
* To use the new SAM 2.1 checkpoints, you need the latest model code from this repo. If you have installed an earlier version of this repo, please first uninstall the previous version via `pip uninstall SAM-2`, pull the latest code from this repo (with `git pull`), and then reinstall the repo following [Installation](#installation) below.
- The training (and fine-tuning) code has been released. See [`training/README.md`](training/README.md) on how to get started.
- The frontend + backend code for the SAM 2 web demo has been released. See [`demo/README.md`](demo/README.md) for details.
## Installation
SAM 2 needs to be installed first before use. The code requires `python>=3.10`, as well as `torch>=2.3.1` and `torchvision>=0.18.1`. Please follow the instructions [here](https://pytorch.org/get-started/locally/) to install both PyTorch and TorchVision dependencies. You can install SAM 2 on a GPU machine using:
```bash
git clone https://github.com/facebookresearch/sam2.git && cd sam2
pip install -e .
```
If you are installing on Windows, it's strongly recommended to use [Windows Subsystem for Linux (WSL)](https://learn.microsoft.com/en-us/windows/wsl/install) with Ubuntu.
To use the SAM 2 predictor and run the example notebooks, `jupyter` and `matplotlib` are required and can be installed by:
```bash
pip install -e ".[notebooks]"
```
Note:
1. It's recommended to create a new Python environment via [Anaconda](https://www.anaconda.com/) for this installation and install PyTorch 2.3.1 (or higher) via `pip` following https://pytorch.org/. If you have a PyTorch version lower than 2.3.1 in your current environment, the installation command above will try to upgrade it to the latest PyTorch version using `pip`.
2. The step above requires compiling a custom CUDA kernel with the `nvcc` compiler. If it isn't already available on your machine, please install the [CUDA toolkits](https://developer.nvidia.com/cuda-toolkit-archive) with a version that matches your PyTorch CUDA version.
3. If you see a message like `Failed to build the SAM 2 CUDA extension` during installation, you can ignore it and still use SAM 2 (some post-processing functionality may be limited, but it doesn't affect the results in most cases).
Please see [`INSTALL.md`](./INSTALL.md) for FAQs on potential issues and solutions.
## Getting Started
### Download Checkpoints
First, we need to download a model checkpoint. All the model checkpoints can be downloaded by running:
```bash
cd checkpoints && \
./download_ckpts.sh && \
cd ..
```
or individually from:
- [sam2.1_hiera_tiny.pt](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_tiny.pt)
- [sam2.1_hiera_small.pt](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_small.pt)
- [sam2.1_hiera_base_plus.pt](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_base_plus.pt)
- [sam2.1_hiera_large.pt](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_large.pt)
(note that these are the improved checkpoints denoted as SAM 2.1; see [Model Description](#model-description) for details.)
Then SAM 2 can be used in a few lines as follows for image and video prediction.
### Image prediction
SAM 2 has all the capabilities of [SAM](https://github.com/facebookresearch/segment-anything) on static images, and we provide image prediction APIs that closely resemble SAM for image use cases. The `SAM2ImagePredictor` class has an easy interface for image prompting.
```python
import torch
from sam2.build_sam import build_sam2
from sam2.sam2_image_predictor import SAM2ImagePredictor
checkpoint = "./checkpoints/sam2.1_hiera_large.pt"
model_cfg = "configs/sam2.1/sam2.1_hiera_l.yaml"
predictor = SAM2ImagePredictor(build_sam2(model_cfg, checkpoint))
with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16):
predictor.set_image(<your_image>)
masks, _, _ = predictor.predict(<input_prompts>)
```
Please refer to the examples in [image_predictor_example.ipynb](./notebooks/image_predictor_example.ipynb) (also in Colab [here](https://colab.research.google.com/github/facebookresearch/sam2/blob/main/notebooks/image_predictor_example.ipynb)) for static image use cases.
SAM 2 also supports automatic mask generation on images just like SAM. Please see [automatic_mask_generator_example.ipynb](./notebooks/automatic_mask_generator_example.ipynb) (also in Colab [here](https://colab.research.google.com/github/facebookresearch/sam2/blob/main/notebooks/automatic_mask_generator_example.ipynb)) for automatic mask generation in images.
### Video prediction
For promptable segmentation and tracking in videos, we provide a video predictor with APIs for example to add prompts and propagate masklets throughout a video. SAM 2 supports video inference on multiple objects and uses an inference state to keep track of the interactions in each video.
```python
import torch
from sam2.build_sam import build_sam2_video_predictor
checkpoint = "./checkpoints/sam2.1_hiera_large.pt"
model_cfg = "configs/sam2.1/sam2.1_hiera_l.yaml"
predictor = build_sam2_video_predictor(model_cfg, checkpoint)
with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16):
state = predictor.init_state(<your_video>)
# add new prompts and instantly get the output on the same frame
frame_idx, object_ids, masks = predictor.add_new_points_or_box(state, <your_prompts>):
# propagate the prompts to get masklets throughout the video
for frame_idx, object_ids, masks in predictor.propagate_in_video(state):
...
```
Please refer to the examples in [video_predictor_example.ipynb](./notebooks/video_predictor_example.ipynb) (also in Colab [here](https://colab.research.google.com/github/facebookresearch/sam2/blob/main/notebooks/video_predictor_example.ipynb)) for details on how to add click or box prompts, make refinements, and track multiple objects in videos.
## Load from 🤗 Hugging Face
Alternatively, models can also be loaded from [Hugging Face](https://huggingface.co/models?search=facebook/sam2) (requires `pip install huggingface_hub`).
For image prediction:
```python
import torch
from sam2.sam2_image_predictor import SAM2ImagePredictor
predictor = SAM2ImagePredictor.from_pretrained("facebook/sam2-hiera-large")
with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16):
predictor.set_image(<your_image>)
masks, _, _ = predictor.predict(<input_prompts>)
```
For video prediction:
```python
import torch
from sam2.sam2_video_predictor import SAM2VideoPredictor
predictor = SAM2VideoPredictor.from_pretrained("facebook/sam2-hiera-large")
with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16):
state = predictor.init_state(<your_video>)
# add new prompts and instantly get the output on the same frame
frame_idx, object_ids, masks = predictor.add_new_points_or_box(state, <your_prompts>):
# propagate the prompts to get masklets throughout the video
for frame_idx, object_ids, masks in predictor.propagate_in_video(state):
...
```
## Model Description
### SAM 2.1 checkpoints
The table below shows the improved SAM 2.1 checkpoints released on September 29, 2024.
| **Model** | **Size (M)** | **Speed (FPS)** | **SA-V test (J&F)** | **MOSE val (J&F)** | **LVOS v2 (J&F)** |
| :------------------: | :----------: | :--------------------: | :-----------------: | :----------------: | :---------------: |
| sam2.1_hiera_tiny <br /> ([config](sam2/configs/sam2.1/sam2.1_hiera_t.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_tiny.pt)) | 38.9 | 47.2 | 76.5 | 71.8 | 77.3 |
| sam2.1_hiera_small <br /> ([config](sam2/configs/sam2.1/sam2.1_hiera_s.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_small.pt)) | 46 | 43.3 (53.0 compiled\*) | 76.6 | 73.5 | 78.3 |
| sam2.1_hiera_base_plus <br /> ([config](sam2/configs/sam2.1/sam2.1_hiera_b+.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_base_plus.pt)) | 80.8 | 34.8 (43.8 compiled\*) | 78.2 | 73.7 | 78.2 |
| sam2.1_hiera_large <br /> ([config](sam2/configs/sam2.1/sam2.1_hiera_l.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_large.pt)) | 224.4 | 24.2 (30.2 compiled\*) | 79.5 | 74.6 | 80.6 |
### SAM 2 checkpoints
The previous SAM 2 checkpoints released on July 29, 2024 can be found as follows:
| **Model** | **Size (M)** | **Speed (FPS)** | **SA-V test (J&F)** | **MOSE val (J&F)** | **LVOS v2 (J&F)** |
| :------------------: | :----------: | :--------------------: | :-----------------: | :----------------: | :---------------: |
| sam2_hiera_tiny <br /> ([config](sam2/configs/sam2/sam2_hiera_t.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_tiny.pt)) | 38.9 | 47.2 | 75.0 | 70.9 | 75.3 |
| sam2_hiera_small <br /> ([config](sam2/configs/sam2/sam2_hiera_s.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_small.pt)) | 46 | 43.3 (53.0 compiled\*) | 74.9 | 71.5 | 76.4 |
| sam2_hiera_base_plus <br /> ([config](sam2/configs/sam2/sam2_hiera_b+.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_base_plus.pt)) | 80.8 | 34.8 (43.8 compiled\*) | 74.7 | 72.8 | 75.8 |
| sam2_hiera_large <br /> ([config](sam2/configs/sam2/sam2_hiera_l.yaml), [checkpoint](https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_large.pt)) | 224.4 | 24.2 (30.2 compiled\*) | 76.0 | 74.6 | 79.8 |
\* Compile the model by setting `compile_image_encoder: True` in the config.
## Segment Anything Video Dataset
See [sav_dataset/README.md](sav_dataset/README.md) for details.
## Training SAM 2
You can train or fine-tune SAM 2 on custom datasets of images, videos, or both. Please check the training [README](training/README.md) on how to get started.
## Web demo for SAM 2
We have released the frontend + backend code for the SAM 2 web demo (a locally deployable version similar to https://sam2.metademolab.com/demo). Please see the web demo [README](demo/README.md) for details.
## License
The SAM 2 model checkpoints, SAM 2 demo code (front-end and back-end), and SAM 2 training code are licensed under [Apache 2.0](./LICENSE), however the [Inter Font](https://github.com/rsms/inter?tab=OFL-1.1-1-ov-file) and [Noto Color Emoji](https://github.com/googlefonts/noto-emoji) used in the SAM 2 demo code are made available under the [SIL Open Font License, version 1.1](https://openfontlicense.org/open-font-license-official-text/).
## Contributing
See [contributing](CONTRIBUTING.md) and the [code of conduct](CODE_OF_CONDUCT.md).
## Contributors
The SAM 2 project was made possible with the help of many contributors (alphabetical):
Karen Bergan, Daniel Bolya, Alex Bosenberg, Kai Brown, Vispi Cassod, Christopher Chedeau, Ida Cheng, Luc Dahlin, Shoubhik Debnath, Rene Martinez Doehner, Grant Gardner, Sahir Gomez, Rishi Godugu, Baishan Guo, Caleb Ho, Andrew Huang, Somya Jain, Bob Kamma, Amanda Kallet, Jake Kinney, Alexander Kirillov, Shiva Koduvayur, Devansh Kukreja, Robert Kuo, Aohan Lin, Parth Malani, Jitendra Malik, Mallika Malhotra, Miguel Martin, Alexander Miller, Sasha Mitts, William Ngan, George Orlin, Joelle Pineau, Kate Saenko, Rodrick Shepard, Azita Shokrpour, David Soofian, Jonathan Torres, Jenny Truong, Sagar Vaze, Meng Wang, Claudette Ward, Pengchuan Zhang.
Third-party code: we use a GPU-based connected component algorithm adapted from [`cc_torch`](https://github.com/zsef123/Connected_components_PyTorch) (with its license in [`LICENSE_cctorch`](./LICENSE_cctorch)) as an optional post-processing step for the mask predictions.
## Citing SAM 2
If you use SAM 2 or the SA-V dataset in your research, please use the following BibTeX entry.
```bibtex
@article{ravi2024sam2,
title={SAM 2: Segment Anything in Images and Videos},
author={Ravi, Nikhila and Gabeur, Valentin and Hu, Yuan-Ting and Hu, Ronghang and Ryali, Chaitanya and Ma, Tengyu and Khedr, Haitham and R{\"a}dle, Roman and Rolland, Chloe and Gustafson, Laura and Mintun, Eric and Pan, Junting and Alwala, Kalyan Vasudev and Carion, Nicolas and Wu, Chao-Yuan and Girshick, Ross and Doll{\'a}r, Piotr and Feichtenhofer, Christoph},
journal={arXiv preprint arXiv:2408.00714},
url={https://arxiv.org/abs/2408.00714},
year={2024}
}
```

BIN
sam2/assets/model_diagram.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 354 KiB

BIN
sam2/assets/sa_v_dataset.jpg Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 545 KiB

64
sam2/backend.Dockerfile Normal file
View File

@@ -0,0 +1,64 @@
ARG BASE_IMAGE=pytorch/pytorch:2.3.1-cuda12.1-cudnn8-runtime
ARG MODEL_SIZE=base_plus
FROM ${BASE_IMAGE}
# Gunicorn environment variables
ENV GUNICORN_WORKERS=1
ENV GUNICORN_THREADS=2
ENV GUNICORN_PORT=5000
# SAM 2 environment variables
ENV APP_ROOT=/opt/sam2
ENV PYTHONUNBUFFERED=1
ENV SAM2_BUILD_CUDA=0
ENV MODEL_SIZE=${MODEL_SIZE}
# Install system requirements
RUN apt-get update && apt-get install -y --no-install-recommends \
ffmpeg \
libavutil-dev \
libavcodec-dev \
libavformat-dev \
libswscale-dev \
pkg-config \
build-essential \
libffi-dev
COPY setup.py .
COPY README.md .
RUN pip install --upgrade pip setuptools
RUN pip install -e ".[interactive-demo]"
# https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite/issues/69#issuecomment-1826764707
RUN rm /opt/conda/bin/ffmpeg && ln -s /bin/ffmpeg /opt/conda/bin/ffmpeg
# Make app directory. This directory will host all files required for the
# backend and SAM 2 inference files.
RUN mkdir ${APP_ROOT}
# Copy backend server files
COPY demo/backend/server ${APP_ROOT}/server
# Copy SAM 2 inference files
COPY sam2 ${APP_ROOT}/server/sam2
# Download SAM 2.1 checkpoints
ADD https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_tiny.pt ${APP_ROOT}/checkpoints/sam2.1_hiera_tiny.pt
ADD https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_small.pt ${APP_ROOT}/checkpoints/sam2.1_hiera_small.pt
ADD https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_base_plus.pt ${APP_ROOT}/checkpoints/sam2.1_hiera_base_plus.pt
ADD https://dl.fbaipublicfiles.com/segment_anything_2/092824/sam2.1_hiera_large.pt ${APP_ROOT}/checkpoints/sam2.1_hiera_large.pt
WORKDIR ${APP_ROOT}/server
# https://pythonspeed.com/articles/gunicorn-in-docker/
CMD gunicorn --worker-tmp-dir /dev/shm \
--worker-class gthread app:app \
--log-level info \
--access-logfile /dev/stdout \
--log-file /dev/stderr \
--workers ${GUNICORN_WORKERS} \
--threads ${GUNICORN_THREADS} \
--bind 0.0.0.0:${GUNICORN_PORT} \
--timeout 60

59
sam2/checkpoints/download_ckpts.sh Executable file
View File

@@ -0,0 +1,59 @@
#!/bin/bash
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
# Use either wget or curl to download the checkpoints
if command -v wget &> /dev/null; then
CMD="wget"
elif command -v curl &> /dev/null; then
CMD="curl -L -O"
else
echo "Please install wget or curl to download the checkpoints."
exit 1
fi
# Define the URLs for SAM 2 checkpoints
# SAM2_BASE_URL="https://dl.fbaipublicfiles.com/segment_anything_2/072824"
# sam2_hiera_t_url="${SAM2_BASE_URL}/sam2_hiera_tiny.pt"
# sam2_hiera_s_url="${SAM2_BASE_URL}/sam2_hiera_small.pt"
# sam2_hiera_b_plus_url="${SAM2_BASE_URL}/sam2_hiera_base_plus.pt"
# sam2_hiera_l_url="${SAM2_BASE_URL}/sam2_hiera_large.pt"
# Download each of the four checkpoints using wget
# echo "Downloading sam2_hiera_tiny.pt checkpoint..."
# $CMD $sam2_hiera_t_url || { echo "Failed to download checkpoint from $sam2_hiera_t_url"; exit 1; }
# echo "Downloading sam2_hiera_small.pt checkpoint..."
# $CMD $sam2_hiera_s_url || { echo "Failed to download checkpoint from $sam2_hiera_s_url"; exit 1; }
# echo "Downloading sam2_hiera_base_plus.pt checkpoint..."
# $CMD $sam2_hiera_b_plus_url || { echo "Failed to download checkpoint from $sam2_hiera_b_plus_url"; exit 1; }
# echo "Downloading sam2_hiera_large.pt checkpoint..."
# $CMD $sam2_hiera_l_url || { echo "Failed to download checkpoint from $sam2_hiera_l_url"; exit 1; }
# Define the URLs for SAM 2.1 checkpoints
SAM2p1_BASE_URL="https://dl.fbaipublicfiles.com/segment_anything_2/092824"
sam2p1_hiera_t_url="${SAM2p1_BASE_URL}/sam2.1_hiera_tiny.pt"
sam2p1_hiera_s_url="${SAM2p1_BASE_URL}/sam2.1_hiera_small.pt"
sam2p1_hiera_b_plus_url="${SAM2p1_BASE_URL}/sam2.1_hiera_base_plus.pt"
sam2p1_hiera_l_url="${SAM2p1_BASE_URL}/sam2.1_hiera_large.pt"
# SAM 2.1 checkpoints
echo "Downloading sam2.1_hiera_tiny.pt checkpoint..."
$CMD $sam2p1_hiera_t_url || { echo "Failed to download checkpoint from $sam2p1_hiera_t_url"; exit 1; }
echo "Downloading sam2.1_hiera_small.pt checkpoint..."
$CMD $sam2p1_hiera_s_url || { echo "Failed to download checkpoint from $sam2p1_hiera_s_url"; exit 1; }
echo "Downloading sam2.1_hiera_base_plus.pt checkpoint..."
$CMD $sam2p1_hiera_b_plus_url || { echo "Failed to download checkpoint from $sam2p1_hiera_b_plus_url"; exit 1; }
echo "Downloading sam2.1_hiera_large.pt checkpoint..."
$CMD $sam2p1_hiera_l_url || { echo "Failed to download checkpoint from $sam2p1_hiera_l_url"; exit 1; }
echo "All checkpoints are downloaded successfully."

2
sam2/demo/.gitignore vendored Normal file
View File

@@ -0,0 +1,2 @@
data/uploads
data/posters

173
sam2/demo/README.md Normal file
View File

@@ -0,0 +1,173 @@
# SAM 2 Demo
Welcome to the SAM 2 Demo! This project consists of a frontend built with React TypeScript and Vite and a backend service using Python Flask and Strawberry GraphQL. Both components can be run in Docker containers or locally on MPS (Metal Performance Shaders) or CPU. However, running the backend service on MPS or CPU devices may result in significantly slower performance (FPS).
## Prerequisites
Before you begin, ensure you have the following installed on your system:
- Docker and Docker Compose
- [OPTIONAL] Node.js and Yarn for running frontend locally
- [OPTIONAL] Anaconda for running backend locally
### Installing Docker
To install Docker, follow these steps:
1. Go to the [Docker website](https://www.docker.com/get-started)
2. Follow the installation instructions for your operating system.
### [OPTIONAL] Installing Node.js and Yarn
To install Node.js and Yarn, follow these steps:
1. Go to the [Node.js website](https://nodejs.org/en/download/).
2. Follow the installation instructions for your operating system.
3. Once Node.js is installed, open a terminal or command prompt and run the following command to install Yarn:
```
npm install -g yarn
```
### [OPTIONAL] Installing Anaconda
To install Anaconda, follow these steps:
1. Go to the [Anaconda website](https://www.anaconda.com/products/distribution).
2. Follow the installation instructions for your operating system.
## Quick Start
To get both the frontend and backend running quickly using Docker, you can use the following command:
```bash
docker compose up --build
```
> [!WARNING]
> On macOS, Docker containers only support running on CPU. MPS is not supported through Docker. If you want to run the demo backend service on MPS, you will need to run it locally (see "Running the Backend Locally" below).
This will build and start both services. You can access them at:
- **Frontend:** [http://localhost:7262](http://localhost:7262)
- **Backend:** [http://localhost:7263/graphql](http://localhost:7263/graphql)
## Running Backend with MPS Support
MPS (Metal Performance Shaders) is not supported with Docker. To use MPS, you need to run the backend on your local machine.
### Setting Up Your Environment
1. **Create Conda environment**
Create a new Conda environment for this project by running the following command or use your existing conda environment for SAM 2:
```
conda create --name sam2-demo python=3.10 --yes
```
This will create a new environment named `sam2-demo` with Python 3.10 as the interpreter.
2. **Activate the Conda environment:**
```bash
conda activate sam2-demo
```
3. **Install ffmpeg**
```bash
conda install -c conda-forge ffmpeg
```
4. **Install SAM 2 demo dependencies:**
Install project dependencies by running the following command in the SAM 2 checkout root directory:
```bash
pip install -e '.[interactive-demo]'
```
### Running the Backend Locally
Download the SAM 2 checkpoints:
```bash
(cd ./checkpoints && ./download_ckpts.sh)
```
Use the following command to start the backend with MPS support:
```bash
cd demo/backend/server/
```
```bash
PYTORCH_ENABLE_MPS_FALLBACK=1 \
APP_ROOT="$(pwd)/../../../" \
APP_URL=http://localhost:7263 \
MODEL_SIZE=base_plus \
DATA_PATH="$(pwd)/../../data" \
DEFAULT_VIDEO_PATH=gallery/05_default_juggle.mp4 \
gunicorn \
--worker-class gthread app:app \
--workers 1 \
--threads 2 \
--bind 0.0.0.0:7263 \
--timeout 60
```
Options for the `MODEL_SIZE` argument are "tiny", "small", "base_plus" (default), and "large".
> [!WARNING]
> Running the backend service on MPS devices can cause fatal crashes with the Gunicorn worker due to insufficient MPS memory. Try switching to CPU devices by setting the `SAM2_DEMO_FORCE_CPU_DEVICE=1` environment variable.
### Starting the Frontend
If you wish to run the frontend separately (useful for development), follow these steps:
1. **Navigate to demo frontend directory:**
```bash
cd demo/frontend
```
2. **Install dependencies:**
```bash
yarn install
```
3. **Start the development server:**
```bash
yarn dev --port 7262
```
This will start the frontend development server on [http://localhost:7262](http://localhost:7262).
## Docker Tips
- To rebuild the Docker containers (useful if you've made changes to the Dockerfile or dependencies):
```bash
docker compose up --build
```
- To stop the Docker containers:
```bash
docker compose down
```
## Contributing
Contributions are welcome! Please read our contributing guidelines to get started.
## License
See the LICENSE file for details.
---
By following these instructions, you should have a fully functional development environment for both the frontend and backend of the SAM 2 Demo. Happy coding!

140
sam2/demo/backend/server/app.py Normal file
View File

@@ -0,0 +1,140 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
import logging
from typing import Any, Generator
from app_conf import (
GALLERY_PATH,
GALLERY_PREFIX,
POSTERS_PATH,
POSTERS_PREFIX,
UPLOADS_PATH,
UPLOADS_PREFIX,
)
from data.loader import preload_data
from data.schema import schema
from data.store import set_videos
from flask import Flask, make_response, Request, request, Response, send_from_directory
from flask_cors import CORS
from inference.data_types import PropagateDataResponse, PropagateInVideoRequest
from inference.multipart import MultipartResponseBuilder
from inference.predictor import InferenceAPI
from strawberry.flask.views import GraphQLView
logger = logging.getLogger(__name__)
app = Flask(__name__)
cors = CORS(app, supports_credentials=True)
videos = preload_data()
set_videos(videos)
inference_api = InferenceAPI()
@app.route("/healthy")
def healthy() -> Response:
return make_response("OK", 200)
@app.route(f"/{GALLERY_PREFIX}/<path:path>", methods=["GET"])
def send_gallery_video(path: str) -> Response:
try:
return send_from_directory(
GALLERY_PATH,
path,
)
except:
raise ValueError("resource not found")
@app.route(f"/{POSTERS_PREFIX}/<path:path>", methods=["GET"])
def send_poster_image(path: str) -> Response:
try:
return send_from_directory(
POSTERS_PATH,
path,
)
except:
raise ValueError("resource not found")
@app.route(f"/{UPLOADS_PREFIX}/<path:path>", methods=["GET"])
def send_uploaded_video(path: str):
try:
return send_from_directory(
UPLOADS_PATH,
path,
)
except:
raise ValueError("resource not found")
# TOOD: Protect route with ToS permission check
@app.route("/propagate_in_video", methods=["POST"])
def propagate_in_video() -> Response:
data = request.json
args = {
"session_id": data["session_id"],
"start_frame_index": data.get("start_frame_index", 0),
}
boundary = "frame"
frame = gen_track_with_mask_stream(boundary, **args)
return Response(frame, mimetype="multipart/x-savi-stream; boundary=" + boundary)
def gen_track_with_mask_stream(
boundary: str,
session_id: str,
start_frame_index: int,
) -> Generator[bytes, None, None]:
with inference_api.autocast_context():
request = PropagateInVideoRequest(
type="propagate_in_video",
session_id=session_id,
start_frame_index=start_frame_index,
)
for chunk in inference_api.propagate_in_video(request=request):
yield MultipartResponseBuilder.build(
boundary=boundary,
headers={
"Content-Type": "application/json; charset=utf-8",
"Frame-Current": "-1",
# Total frames minus the reference frame
"Frame-Total": "-1",
"Mask-Type": "RLE[]",
},
body=chunk.to_json().encode("UTF-8"),
).get_message()
class MyGraphQLView(GraphQLView):
def get_context(self, request: Request, response: Response) -> Any:
return {"inference_api": inference_api}
# Add GraphQL route to Flask app.
app.add_url_rule(
"/graphql",
view_func=MyGraphQLView.as_view(
"graphql_view",
schema=schema,
# Disable GET queries
# https://strawberry.rocks/docs/operations/deployment
# https://strawberry.rocks/docs/integrations/flask
allow_queries_via_get=False,
# Strawberry recently changed multipart request handling, which now
# requires enabling support explicitly for views.
# https://github.com/strawberry-graphql/strawberry/issues/3655
multipart_uploads_enabled=True,
),
)
if __name__ == "__main__":
app.run(host="0.0.0.0", port=5000)

55
sam2/demo/backend/server/app_conf.py Normal file
View File

@@ -0,0 +1,55 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
import logging
import os
from pathlib import Path
logger = logging.getLogger(__name__)
APP_ROOT = os.getenv("APP_ROOT", "/opt/sam2")
API_URL = os.getenv("API_URL", "http://localhost:7263")
MODEL_SIZE = os.getenv("MODEL_SIZE", "base_plus")
logger.info(f"using model size {MODEL_SIZE}")
FFMPEG_NUM_THREADS = int(os.getenv("FFMPEG_NUM_THREADS", "1"))
# Path for all data used in API
DATA_PATH = Path(os.getenv("DATA_PATH", "/data"))
# Max duration an uploaded video can have in seconds. The default is 10
# seconds.
MAX_UPLOAD_VIDEO_DURATION = float(os.environ.get("MAX_UPLOAD_VIDEO_DURATION", "10"))
# If set, it will define which video is returned by the default video query for
# desktop
DEFAULT_VIDEO_PATH = os.getenv("DEFAULT_VIDEO_PATH")
# Prefix for gallery videos
GALLERY_PREFIX = "gallery"
# Path where all gallery videos are stored
GALLERY_PATH = DATA_PATH / GALLERY_PREFIX
# Prefix for uploaded videos
UPLOADS_PREFIX = "uploads"
# Path where all uploaded videos are stored
UPLOADS_PATH = DATA_PATH / UPLOADS_PREFIX
# Prefix for video posters (1st frame of video)
POSTERS_PREFIX = "posters"
# Path where all posters are stored
POSTERS_PATH = DATA_PATH / POSTERS_PREFIX
# Make sure any of those paths exist
os.makedirs(DATA_PATH, exist_ok=True)
os.makedirs(GALLERY_PATH, exist_ok=True)
os.makedirs(UPLOADS_PATH, exist_ok=True)
os.makedirs(POSTERS_PATH, exist_ok=True)

154
sam2/demo/backend/server/data/data_types.py Normal file
View File

@@ -0,0 +1,154 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
from dataclasses import dataclass
from typing import Iterable, List, Optional
import strawberry
from app_conf import API_URL
from data.resolver import resolve_videos
from dataclasses_json import dataclass_json
from strawberry import relay
@strawberry.type
class Video(relay.Node):
"""Core type for video."""
code: relay.NodeID[str]
path: str
poster_path: Optional[str]
width: int
height: int
@strawberry.field
def url(self) -> str:
return f"{API_URL}/{self.path}"
@strawberry.field
def poster_url(self) -> str:
return f"{API_URL}/{self.poster_path}"
@classmethod
def resolve_nodes(
cls,
*,
info: relay.PageInfo,
node_ids: Iterable[str],
required: bool = False,
):
return resolve_videos(node_ids, required)
@strawberry.type
class RLEMask:
"""Core type for Onevision GraphQL RLE mask."""
size: List[int]
counts: str
order: str
@strawberry.type
class RLEMaskForObject:
"""Type for RLE mask associated with a specific object id."""
object_id: int
rle_mask: RLEMask
@strawberry.type
class RLEMaskListOnFrame:
"""Type for a list of object-associated RLE masks on a specific video frame."""
frame_index: int
rle_mask_list: List[RLEMaskForObject]
@strawberry.input
class StartSessionInput:
path: str
@strawberry.type
class StartSession:
session_id: str
@strawberry.input
class PingInput:
session_id: str
@strawberry.type
class Pong:
success: bool
@strawberry.input
class CloseSessionInput:
session_id: str
@strawberry.type
class CloseSession:
success: bool
@strawberry.input
class AddPointsInput:
session_id: str
frame_index: int
clear_old_points: bool
object_id: int
labels: List[int]
points: List[List[float]]
@strawberry.input
class ClearPointsInFrameInput:
session_id: str
frame_index: int
object_id: int
@strawberry.input
class ClearPointsInVideoInput:
session_id: str
@strawberry.type
class ClearPointsInVideo:
success: bool
@strawberry.input
class RemoveObjectInput:
session_id: str
object_id: int
@strawberry.input
class PropagateInVideoInput:
session_id: str
start_frame_index: int
@strawberry.input
class CancelPropagateInVideoInput:
session_id: str
@strawberry.type
class CancelPropagateInVideo:
success: bool
@strawberry.type
class SessionExpiration:
session_id: str
expiration_time: int
max_expiration_time: int
ttl: int

92
sam2/demo/backend/server/data/loader.py Normal file
View File

@@ -0,0 +1,92 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
import os
import shutil
import subprocess
from glob import glob
from pathlib import Path
from typing import Dict, Optional
import imagesize
from app_conf import GALLERY_PATH, POSTERS_PATH, POSTERS_PREFIX
from data.data_types import Video
from tqdm import tqdm
def preload_data() -> Dict[str, Video]:
"""
Preload data including gallery videos and their posters.
"""
# Dictionaries for videos and datasets on the backend.
# Note that since Python 3.7, dictionaries preserve their insert order, so
# when looping over its `.values()`, elements inserted first also appear first.
# https://stackoverflow.com/questions/39980323/are-dictionaries-ordered-in-python-3-6
all_videos = {}
video_path_pattern = os.path.join(GALLERY_PATH, "**/*.mp4")
video_paths = glob(video_path_pattern, recursive=True)
for p in tqdm(video_paths):
video = get_video(p, GALLERY_PATH)
all_videos[video.code] = video
return all_videos
def get_video(
filepath: os.PathLike,
absolute_path: Path,
file_key: Optional[str] = None,
generate_poster: bool = True,
width: Optional[int] = None,
height: Optional[int] = None,
verbose: Optional[bool] = False,
) -> Video:
"""
Get video object given
"""
# Use absolute_path to include the parent directory in the video
video_path = os.path.relpath(filepath, absolute_path.parent)
poster_path = None
if generate_poster:
poster_id = os.path.splitext(os.path.basename(filepath))[0]
poster_filename = f"{str(poster_id)}.jpg"
poster_path = f"{POSTERS_PREFIX}/{poster_filename}"
# Extract the first frame from video
poster_output_path = os.path.join(POSTERS_PATH, poster_filename)
ffmpeg = shutil.which("ffmpeg")
subprocess.call(
[
ffmpeg,
"-y",
"-i",
str(filepath),
"-pix_fmt",
"yuv420p",
"-frames:v",
"1",
"-update",
"1",
"-strict",
"unofficial",
str(poster_output_path),
],
stdout=None if verbose else subprocess.DEVNULL,
stderr=None if verbose else subprocess.DEVNULL,
)
# Extract video width and height from poster. This is important to optimize
# rendering previews in the mosaic video preview.
width, height = imagesize.get(poster_output_path)
return Video(
code=video_path,
path=video_path if file_key is None else file_key,
poster_path=poster_path,
width=width,
height=height,
)

18
sam2/demo/backend/server/data/resolver.py Normal file
View File

@@ -0,0 +1,18 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
from typing import Iterable
def resolve_videos(node_ids: Iterable[str], required: bool = False):
"""
Resolve videos given node ids.
"""
from data.store import get_videos
all_videos = get_videos()
return [
all_videos[nid] if required else all_videos.get(nid, None) for nid in node_ids
]

357
sam2/demo/backend/server/data/schema.py Normal file
View File

@@ -0,0 +1,357 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
import hashlib
import os
import shutil
import tempfile
from pathlib import Path
from typing import Iterable, List, Optional, Tuple, Union
import av
import strawberry
from app_conf import (
DATA_PATH,
DEFAULT_VIDEO_PATH,
MAX_UPLOAD_VIDEO_DURATION,
UPLOADS_PATH,
UPLOADS_PREFIX,
)
from data.data_types import (
AddPointsInput,
CancelPropagateInVideo,
CancelPropagateInVideoInput,
ClearPointsInFrameInput,
ClearPointsInVideo,
ClearPointsInVideoInput,
CloseSession,
CloseSessionInput,
RemoveObjectInput,
RLEMask,
RLEMaskForObject,
RLEMaskListOnFrame,
StartSession,
StartSessionInput,
Video,
)
from data.loader import get_video
from data.store import get_videos
from data.transcoder import get_video_metadata, transcode, VideoMetadata
from inference.data_types import (
AddPointsRequest,
CancelPropagateInVideoRequest,
CancelPropagateInVideoRequest,
ClearPointsInFrameRequest,
ClearPointsInVideoRequest,
CloseSessionRequest,
RemoveObjectRequest,
StartSessionRequest,
)
from inference.predictor import InferenceAPI
from strawberry import relay
from strawberry.file_uploads import Upload
@strawberry.type
class Query:
@strawberry.field
def default_video(self) -> Video:
"""
Return the default video.
The default video can be set with the DEFAULT_VIDEO_PATH environment
variable. It will return the video that matches this path. If no video
is found, it will return the first video.
"""
all_videos = get_videos()
# Find the video that matches the default path and return that as
# default video.
for _, v in all_videos.items():
if v.path == DEFAULT_VIDEO_PATH:
return v
# Fallback is returning the first video
return next(iter(all_videos.values()))
@relay.connection(relay.ListConnection[Video])
def videos(
self,
) -> Iterable[Video]:
"""
Return all available videos.
"""
all_videos = get_videos()
return all_videos.values()
@strawberry.type
class Mutation:
@strawberry.mutation
def upload_video(
self,
file: Upload,
start_time_sec: Optional[float] = None,
duration_time_sec: Optional[float] = None,
) -> Video:
"""
Receive a video file and store it in the configured S3 bucket.
"""
max_time = MAX_UPLOAD_VIDEO_DURATION
filepath, file_key, vm = process_video(
file,
max_time=max_time,
start_time_sec=start_time_sec,
duration_time_sec=duration_time_sec,
)
video = get_video(
filepath,
UPLOADS_PATH,
file_key=file_key,
width=vm.width,
height=vm.height,
generate_poster=False,
)
return video
@strawberry.mutation
def start_session(
self, input: StartSessionInput, info: strawberry.Info
) -> StartSession:
inference_api: InferenceAPI = info.context["inference_api"]
request = StartSessionRequest(
type="start_session",
path=f"{DATA_PATH}/{input.path}",
)
response = inference_api.start_session(request=request)
return StartSession(session_id=response.session_id)
@strawberry.mutation
def close_session(
self, input: CloseSessionInput, info: strawberry.Info
) -> CloseSession:
inference_api: InferenceAPI = info.context["inference_api"]
request = CloseSessionRequest(
type="close_session",
session_id=input.session_id,
)
response = inference_api.close_session(request)
return CloseSession(success=response.success)
@strawberry.mutation
def add_points(
self, input: AddPointsInput, info: strawberry.Info
) -> RLEMaskListOnFrame:
inference_api: InferenceAPI = info.context["inference_api"]
request = AddPointsRequest(
type="add_points",
session_id=input.session_id,
frame_index=input.frame_index,
object_id=input.object_id,
points=input.points,
labels=input.labels,
clear_old_points=input.clear_old_points,
)
reponse = inference_api.add_points(request)
return RLEMaskListOnFrame(
frame_index=reponse.frame_index,
rle_mask_list=[
RLEMaskForObject(
object_id=r.object_id,
rle_mask=RLEMask(counts=r.mask.counts, size=r.mask.size, order="F"),
)
for r in reponse.results
],
)
@strawberry.mutation
def remove_object(
self, input: RemoveObjectInput, info: strawberry.Info
) -> List[RLEMaskListOnFrame]:
inference_api: InferenceAPI = info.context["inference_api"]
request = RemoveObjectRequest(
type="remove_object", session_id=input.session_id, object_id=input.object_id
)
response = inference_api.remove_object(request)
return [
RLEMaskListOnFrame(
frame_index=res.frame_index,
rle_mask_list=[
RLEMaskForObject(
object_id=r.object_id,
rle_mask=RLEMask(
counts=r.mask.counts, size=r.mask.size, order="F"
),
)
for r in res.results
],
)
for res in response.results
]
@strawberry.mutation
def clear_points_in_frame(
self, input: ClearPointsInFrameInput, info: strawberry.Info
) -> RLEMaskListOnFrame:
inference_api: InferenceAPI = info.context["inference_api"]
request = ClearPointsInFrameRequest(
type="clear_points_in_frame",
session_id=input.session_id,
frame_index=input.frame_index,
object_id=input.object_id,
)
response = inference_api.clear_points_in_frame(request)
return RLEMaskListOnFrame(
frame_index=response.frame_index,
rle_mask_list=[
RLEMaskForObject(
object_id=r.object_id,
rle_mask=RLEMask(counts=r.mask.counts, size=r.mask.size, order="F"),
)
for r in response.results
],
)
@strawberry.mutation
def clear_points_in_video(
self, input: ClearPointsInVideoInput, info: strawberry.Info
) -> ClearPointsInVideo:
inference_api: InferenceAPI = info.context["inference_api"]
request = ClearPointsInVideoRequest(
type="clear_points_in_video",
session_id=input.session_id,
)
response = inference_api.clear_points_in_video(request)
return ClearPointsInVideo(success=response.success)
@strawberry.mutation
def cancel_propagate_in_video(
self, input: CancelPropagateInVideoInput, info: strawberry.Info
) -> CancelPropagateInVideo:
inference_api: InferenceAPI = info.context["inference_api"]
request = CancelPropagateInVideoRequest(
type="cancel_propagate_in_video",
session_id=input.session_id,
)
response = inference_api.cancel_propagate_in_video(request)
return CancelPropagateInVideo(success=response.success)
def get_file_hash(video_path_or_file) -> str:
if isinstance(video_path_or_file, str):
with open(video_path_or_file, "rb") as in_f:
result = hashlib.sha256(in_f.read()).hexdigest()
else:
video_path_or_file.seek(0)
result = hashlib.sha256(video_path_or_file.read()).hexdigest()
return result
def _get_start_sec_duration_sec(
start_time_sec: Union[float, None],
duration_time_sec: Union[float, None],
max_time: float,
) -> Tuple[float, float]:
default_seek_t = int(os.environ.get("VIDEO_ENCODE_SEEK_TIME", "0"))
if start_time_sec is None:
start_time_sec = default_seek_t
if duration_time_sec is not None:
duration_time_sec = min(duration_time_sec, max_time)
else:
duration_time_sec = max_time
return start_time_sec, duration_time_sec
def process_video(
file: Upload,
max_time: float,
start_time_sec: Optional[float] = None,
duration_time_sec: Optional[float] = None,
) -> Tuple[Optional[str], str, str, VideoMetadata]:
"""
Process file upload including video trimming and content moderation checks.
Returns the filepath, s3_file_key, hash & video metaedata as a tuple.
"""
with tempfile.TemporaryDirectory() as tempdir:
in_path = f"{tempdir}/in.mp4"
out_path = f"{tempdir}/out.mp4"
with open(in_path, "wb") as in_f:
in_f.write(file.read())
try:
video_metadata = get_video_metadata(in_path)
except av.InvalidDataError:
raise Exception("not valid video file")
if video_metadata.num_video_streams == 0:
raise Exception("video container does not contain a video stream")
if video_metadata.width is None or video_metadata.height is None:
raise Exception("video container does not contain width or height metadata")
if video_metadata.duration_sec in (None, 0):
raise Exception("video container does time duration metadata")
start_time_sec, duration_time_sec = _get_start_sec_duration_sec(
max_time=max_time,
start_time_sec=start_time_sec,
duration_time_sec=duration_time_sec,
)
# Transcode video to make sure videos returned to the app are all in
# the same format, duration, resolution, fps.
transcode(
in_path,
out_path,
video_metadata,
seek_t=start_time_sec,
duration_time_sec=duration_time_sec,
)
os.remove(in_path) # don't need original video now
out_video_metadata = get_video_metadata(out_path)
if out_video_metadata.num_video_frames == 0:
raise Exception(
"transcode produced empty video; check seek time or your input video"
)
filepath = None
file_key = None
with open(out_path, "rb") as file_data:
file_hash = get_file_hash(file_data)
file_data.seek(0)
file_key = UPLOADS_PREFIX + "/" + f"{file_hash}.mp4"
filepath = os.path.join(UPLOADS_PATH, f"{file_hash}.mp4")
assert filepath is not None and file_key is not None
shutil.move(out_path, filepath)
return filepath, file_key, out_video_metadata
schema = strawberry.Schema(
query=Query,
mutation=Mutation,
)

28
sam2/demo/backend/server/data/store.py Normal file
View File

@@ -0,0 +1,28 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
from typing import Dict
from data.data_types import Video
ALL_VIDEOS: Dict[str, Video] = []
def set_videos(videos: Dict[str, Video]) -> None:
"""
Set the videos available in the backend. The data is kept in-memory, but a future change could replace the
in-memory storage with a database backend. This would also be more efficient when querying videos given a
dataset name etc.
"""
global ALL_VIDEOS
ALL_VIDEOS = videos
def get_videos() -> Dict[str, Video]:
"""
Return the videos available in the backend.
"""
global ALL_VIDEOS
return ALL_VIDEOS

186
sam2/demo/backend/server/data/transcoder.py Normal file
View File

@@ -0,0 +1,186 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
import ast
import math
import os
import shutil
import subprocess
from dataclasses import dataclass
from typing import Optional
import av
from app_conf import FFMPEG_NUM_THREADS
from dataclasses_json import dataclass_json
TRANSCODE_VERSION = 1
@dataclass_json
@dataclass
class VideoMetadata:
duration_sec: Optional[float]
video_duration_sec: Optional[float]
container_duration_sec: Optional[float]
fps: Optional[float]
width: Optional[int]
height: Optional[int]
num_video_frames: int
num_video_streams: int
video_start_time: float
def transcode(
in_path: str,
out_path: str,
in_metadata: Optional[VideoMetadata],
seek_t: float,
duration_time_sec: float,
):
codec = os.environ.get("VIDEO_ENCODE_CODEC", "libx264")
crf = int(os.environ.get("VIDEO_ENCODE_CRF", "23"))
fps = int(os.environ.get("VIDEO_ENCODE_FPS", "24"))
max_w = int(os.environ.get("VIDEO_ENCODE_MAX_WIDTH", "1280"))
max_h = int(os.environ.get("VIDEO_ENCODE_MAX_HEIGHT", "720"))
verbose = ast.literal_eval(os.environ.get("VIDEO_ENCODE_VERBOSE", "False"))
normalize_video(
in_path=in_path,
out_path=out_path,
max_w=max_w,
max_h=max_h,
seek_t=seek_t,
max_time=duration_time_sec,
in_metadata=in_metadata,
codec=codec,
crf=crf,
fps=fps,
verbose=verbose,
)
def get_video_metadata(path: str) -> VideoMetadata:
with av.open(path) as cont:
num_video_streams = len(cont.streams.video)
width, height, fps = None, None, None
video_duration_sec = 0
container_duration_sec = float((cont.duration or 0) / av.time_base)
video_start_time = 0.0
rotation_deg = 0
num_video_frames = 0
if num_video_streams > 0:
video_stream = cont.streams.video[0]
assert video_stream.time_base is not None
# for rotation, see: https://github.com/PyAV-Org/PyAV/pull/1249
rotation_deg = video_stream.side_data.get("DISPLAYMATRIX", 0)
num_video_frames = video_stream.frames
video_start_time = float(video_stream.start_time * video_stream.time_base)
width, height = video_stream.width, video_stream.height
fps = float(video_stream.guessed_rate)
fps_avg = video_stream.average_rate
if video_stream.duration is not None:
video_duration_sec = float(
video_stream.duration * video_stream.time_base
)
if fps is None:
fps = float(fps_avg)
if not math.isnan(rotation_deg) and int(rotation_deg) in (
90,
-90,
270,
-270,
):
width, height = height, width
duration_sec = max(container_duration_sec, video_duration_sec)
return VideoMetadata(
duration_sec=duration_sec,
container_duration_sec=container_duration_sec,
video_duration_sec=video_duration_sec,
video_start_time=video_start_time,
fps=fps,
width=width,
height=height,
num_video_streams=num_video_streams,
num_video_frames=num_video_frames,
)
def normalize_video(
in_path: str,
out_path: str,
max_w: int,
max_h: int,
seek_t: float,
max_time: float,
in_metadata: Optional[VideoMetadata],
codec: str = "libx264",
crf: int = 23,
fps: int = 24,
verbose: bool = False,
):
if in_metadata is None:
in_metadata = get_video_metadata(in_path)
assert in_metadata.num_video_streams > 0, "no video stream present"
w, h = in_metadata.width, in_metadata.height
assert w is not None, "width not available"
assert h is not None, "height not available"
# rescale to max_w:max_h if needed & preserve aspect ratio
r = w / h
if r < 1:
h = min(720, h)
w = h * r
else:
w = min(1280, w)
h = w / r
# h264 cannot encode w/ odd dimensions
w = int(w)
h = int(h)
if w % 2 != 0:
w += 1
if h % 2 != 0:
h += 1
ffmpeg = shutil.which("ffmpeg")
cmd = [
ffmpeg,
"-threads",
f"{FFMPEG_NUM_THREADS}", # global threads
"-ss",
f"{seek_t:.2f}",
"-t",
f"{max_time:.2f}",
"-i",
in_path,
"-threads",
f"{FFMPEG_NUM_THREADS}", # decode (or filter..?) threads
"-vf",
f"fps={fps},scale={w}:{h},setsar=1:1",
"-c:v",
codec,
"-crf",
f"{crf}",
"-pix_fmt",
"yuv420p",
"-threads",
f"{FFMPEG_NUM_THREADS}", # encode threads
out_path,
"-y",
]
if verbose:
print(" ".join(cmd))
subprocess.call(
cmd,
stdout=None if verbose else subprocess.DEVNULL,
stderr=None if verbose else subprocess.DEVNULL,
)

191
sam2/demo/backend/server/inference/data_types.py Normal file
View File

@@ -0,0 +1,191 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
from dataclasses import dataclass
from typing import Dict, List, Optional, Union
from dataclasses_json import dataclass_json
from torch import Tensor
@dataclass_json
@dataclass
class Mask:
size: List[int]
counts: str
@dataclass_json
@dataclass
class BaseRequest:
type: str
@dataclass_json
@dataclass
class StartSessionRequest(BaseRequest):
type: str
path: str
session_id: Optional[str] = None
@dataclass_json
@dataclass
class SaveSessionRequest(BaseRequest):
type: str
session_id: str
@dataclass_json
@dataclass
class LoadSessionRequest(BaseRequest):
type: str
session_id: str
@dataclass_json
@dataclass
class RenewSessionRequest(BaseRequest):
type: str
session_id: str
@dataclass_json
@dataclass
class CloseSessionRequest(BaseRequest):
type: str
session_id: str
@dataclass_json
@dataclass
class AddPointsRequest(BaseRequest):
type: str
session_id: str
frame_index: int
clear_old_points: bool
object_id: int
labels: List[int]
points: List[List[float]]
@dataclass_json
@dataclass
class AddMaskRequest(BaseRequest):
type: str
session_id: str
frame_index: int
object_id: int
mask: Mask
@dataclass_json
@dataclass
class ClearPointsInFrameRequest(BaseRequest):
type: str
session_id: str
frame_index: int
object_id: int
@dataclass_json
@dataclass
class ClearPointsInVideoRequest(BaseRequest):
type: str
session_id: str
@dataclass_json
@dataclass
class RemoveObjectRequest(BaseRequest):
type: str
session_id: str
object_id: int
@dataclass_json
@dataclass
class PropagateInVideoRequest(BaseRequest):
type: str
session_id: str
start_frame_index: int
@dataclass_json
@dataclass
class CancelPropagateInVideoRequest(BaseRequest):
type: str
session_id: str
@dataclass_json
@dataclass
class StartSessionResponse:
session_id: str
@dataclass_json
@dataclass
class SaveSessionResponse:
session_id: str
@dataclass_json
@dataclass
class LoadSessionResponse:
session_id: str
@dataclass_json
@dataclass
class RenewSessionResponse:
session_id: str
@dataclass_json
@dataclass
class CloseSessionResponse:
success: bool
@dataclass_json
@dataclass
class ClearPointsInVideoResponse:
success: bool
@dataclass_json
@dataclass
class PropagateDataValue:
object_id: int
mask: Mask
@dataclass_json
@dataclass
class PropagateDataResponse:
frame_index: int
results: List[PropagateDataValue]
@dataclass_json
@dataclass
class RemoveObjectResponse:
results: List[PropagateDataResponse]
@dataclass_json
@dataclass
class CancelPorpagateResponse:
success: bool
@dataclass_json
@dataclass
class InferenceSession:
start_time: float
last_use_time: float
session_id: str
state: Dict[str, Dict[str, Union[Tensor, Dict[int, Tensor]]]]

48
sam2/demo/backend/server/inference/multipart.py Normal file
View File

@@ -0,0 +1,48 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
from typing import Dict, Union
class MultipartResponseBuilder:
message: bytes
def __init__(self, boundary: str) -> None:
self.message = b"--" + boundary.encode("utf-8") + b"\r\n"
@classmethod
def build(
cls, boundary: str, headers: Dict[str, str], body: Union[str, bytes]
) -> "MultipartResponseBuilder":
builder = cls(boundary=boundary)
for k, v in headers.items():
builder.__append_header(key=k, value=v)
if isinstance(body, bytes):
builder.__append_body(body)
elif isinstance(body, str):
builder.__append_body(body.encode("utf-8"))
else:
raise ValueError(
f"body needs to be of type bytes or str but got {type(body)}"
)
return builder
def get_message(self) -> bytes:
return self.message
def __append_header(self, key: str, value: str) -> "MultipartResponseBuilder":
self.message += key.encode("utf-8") + b": " + value.encode("utf-8") + b"\r\n"
return self
def __close_header(self) -> "MultipartResponseBuilder":
self.message += b"\r\n"
return self
def __append_body(self, body: bytes) -> "MultipartResponseBuilder":
self.__append_header(key="Content-Length", value=str(len(body)))
self.__close_header()
self.message += body
return self

427
sam2/demo/backend/server/inference/predictor.py Normal file
View File

@@ -0,0 +1,427 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
import contextlib
import logging
import os
import uuid
from pathlib import Path
from threading import Lock
from typing import Any, Dict, Generator, List
import numpy as np
import torch
from app_conf import APP_ROOT, MODEL_SIZE
from inference.data_types import (
AddMaskRequest,
AddPointsRequest,
CancelPorpagateResponse,
CancelPropagateInVideoRequest,
ClearPointsInFrameRequest,
ClearPointsInVideoRequest,
ClearPointsInVideoResponse,
CloseSessionRequest,
CloseSessionResponse,
Mask,
PropagateDataResponse,
PropagateDataValue,
PropagateInVideoRequest,
RemoveObjectRequest,
RemoveObjectResponse,
StartSessionRequest,
StartSessionResponse,
)
from pycocotools.mask import decode as decode_masks, encode as encode_masks
from sam2.build_sam import build_sam2_video_predictor
logger = logging.getLogger(__name__)
class InferenceAPI:
def __init__(self) -> None:
super(InferenceAPI, self).__init__()
self.session_states: Dict[str, Any] = {}
self.score_thresh = 0
if MODEL_SIZE == "tiny":
checkpoint = Path(APP_ROOT) / "checkpoints/sam2.1_hiera_tiny.pt"
model_cfg = "configs/sam2.1/sam2.1_hiera_t.yaml"
elif MODEL_SIZE == "small":
checkpoint = Path(APP_ROOT) / "checkpoints/sam2.1_hiera_small.pt"
model_cfg = "configs/sam2.1/sam2.1_hiera_s.yaml"
elif MODEL_SIZE == "large":
checkpoint = Path(APP_ROOT) / "checkpoints/sam2.1_hiera_large.pt"
model_cfg = "configs/sam2.1/sam2.1_hiera_l.yaml"
else: # base_plus (default)
checkpoint = Path(APP_ROOT) / "checkpoints/sam2.1_hiera_base_plus.pt"
model_cfg = "configs/sam2.1/sam2.1_hiera_b+.yaml"
# select the device for computation
force_cpu_device = os.environ.get("SAM2_DEMO_FORCE_CPU_DEVICE", "0") == "1"
if force_cpu_device:
logger.info("forcing CPU device for SAM 2 demo")
if torch.cuda.is_available() and not force_cpu_device:
device = torch.device("cuda")
elif torch.backends.mps.is_available() and not force_cpu_device:
device = torch.device("mps")
else:
device = torch.device("cpu")
logger.info(f"using device: {device}")
if device.type == "cuda":
# turn on tfloat32 for Ampere GPUs (https://pytorch.org/docs/stable/notes/cuda.html#tensorfloat-32-tf32-on-ampere-devices)
if torch.cuda.get_device_properties(0).major >= 8:
torch.backends.cuda.matmul.allow_tf32 = True
torch.backends.cudnn.allow_tf32 = True
elif device.type == "mps":
logging.warning(
"\nSupport for MPS devices is preliminary. SAM 2 is trained with CUDA and might "
"give numerically different outputs and sometimes degraded performance on MPS. "
"See e.g. https://github.com/pytorch/pytorch/issues/84936 for a discussion."
)
self.device = device
self.predictor = build_sam2_video_predictor(
model_cfg, checkpoint, device=device
)
self.inference_lock = Lock()
def autocast_context(self):
if self.device.type == "cuda":
return torch.autocast("cuda", dtype=torch.bfloat16)
else:
return contextlib.nullcontext()
def start_session(self, request: StartSessionRequest) -> StartSessionResponse:
with self.autocast_context(), self.inference_lock:
session_id = str(uuid.uuid4())
# for MPS devices, we offload the video frames to CPU by default to avoid
# memory fragmentation in MPS (which sometimes crashes the entire process)
offload_video_to_cpu = self.device.type == "mps"
inference_state = self.predictor.init_state(
request.path,
offload_video_to_cpu=offload_video_to_cpu,
)
self.session_states[session_id] = {
"canceled": False,
"state": inference_state,
}
return StartSessionResponse(session_id=session_id)
def close_session(self, request: CloseSessionRequest) -> CloseSessionResponse:
is_successful = self.__clear_session_state(request.session_id)
return CloseSessionResponse(success=is_successful)
def add_points(
self, request: AddPointsRequest, test: str = ""
) -> PropagateDataResponse:
with self.autocast_context(), self.inference_lock:
session = self.__get_session(request.session_id)
inference_state = session["state"]
frame_idx = request.frame_index
obj_id = request.object_id
points = request.points
labels = request.labels
clear_old_points = request.clear_old_points
# add new prompts and instantly get the output on the same frame
frame_idx, object_ids, masks = self.predictor.add_new_points_or_box(
inference_state=inference_state,
frame_idx=frame_idx,
obj_id=obj_id,
points=points,
labels=labels,
clear_old_points=clear_old_points,
normalize_coords=False,
)
masks_binary = (masks > self.score_thresh)[:, 0].cpu().numpy()
rle_mask_list = self.__get_rle_mask_list(
object_ids=object_ids, masks=masks_binary
)
return PropagateDataResponse(
frame_index=frame_idx,
results=rle_mask_list,
)
def add_mask(self, request: AddMaskRequest) -> PropagateDataResponse:
"""
Add new points on a specific video frame.
- mask is a numpy array of shape [H_im, W_im] (containing 1 for foreground and 0 for background).
Note: providing an input mask would overwrite any previous input points on this frame.
"""
with self.autocast_context(), self.inference_lock:
session_id = request.session_id
frame_idx = request.frame_index
obj_id = request.object_id
rle_mask = {
"counts": request.mask.counts,
"size": request.mask.size,
}
mask = decode_masks(rle_mask)
logger.info(
f"add mask on frame {frame_idx} in session {session_id}: {obj_id=}, {mask.shape=}"
)
session = self.__get_session(session_id)
inference_state = session["state"]
frame_idx, obj_ids, video_res_masks = self.model.add_new_mask(
inference_state=inference_state,
frame_idx=frame_idx,
obj_id=obj_id,
mask=torch.tensor(mask > 0),
)
masks_binary = (video_res_masks > self.score_thresh)[:, 0].cpu().numpy()
rle_mask_list = self.__get_rle_mask_list(
object_ids=obj_ids, masks=masks_binary
)
return PropagateDataResponse(
frame_index=frame_idx,
results=rle_mask_list,
)
def clear_points_in_frame(
self, request: ClearPointsInFrameRequest
) -> PropagateDataResponse:
"""
Remove all input points in a specific frame.
"""
with self.autocast_context(), self.inference_lock:
session_id = request.session_id
frame_idx = request.frame_index
obj_id = request.object_id
logger.info(
f"clear inputs on frame {frame_idx} in session {session_id}: {obj_id=}"
)
session = self.__get_session(session_id)
inference_state = session["state"]
frame_idx, obj_ids, video_res_masks = (
self.predictor.clear_all_prompts_in_frame(
inference_state, frame_idx, obj_id
)
)
masks_binary = (video_res_masks > self.score_thresh)[:, 0].cpu().numpy()
rle_mask_list = self.__get_rle_mask_list(
object_ids=obj_ids, masks=masks_binary
)
return PropagateDataResponse(
frame_index=frame_idx,
results=rle_mask_list,
)
def clear_points_in_video(
self, request: ClearPointsInVideoRequest
) -> ClearPointsInVideoResponse:
"""
Remove all input points in all frames throughout the video.
"""
with self.autocast_context(), self.inference_lock:
session_id = request.session_id
logger.info(f"clear all inputs across the video in session {session_id}")
session = self.__get_session(session_id)
inference_state = session["state"]
self.predictor.reset_state(inference_state)
return ClearPointsInVideoResponse(success=True)
def remove_object(self, request: RemoveObjectRequest) -> RemoveObjectResponse:
"""
Remove an object id from the tracking state.
"""
with self.autocast_context(), self.inference_lock:
session_id = request.session_id
obj_id = request.object_id
logger.info(f"remove object in session {session_id}: {obj_id=}")
session = self.__get_session(session_id)
inference_state = session["state"]
new_obj_ids, updated_frames = self.predictor.remove_object(
inference_state, obj_id
)
results = []
for frame_index, video_res_masks in updated_frames:
masks = (video_res_masks > self.score_thresh)[:, 0].cpu().numpy()
rle_mask_list = self.__get_rle_mask_list(
object_ids=new_obj_ids, masks=masks
)
results.append(
PropagateDataResponse(
frame_index=frame_index,
results=rle_mask_list,
)
)
return RemoveObjectResponse(results=results)
def propagate_in_video(
self, request: PropagateInVideoRequest
) -> Generator[PropagateDataResponse, None, None]:
session_id = request.session_id
start_frame_idx = request.start_frame_index
propagation_direction = "both"
max_frame_num_to_track = None
"""
Propagate existing input points in all frames to track the object across video.
"""
# Note that as this method is a generator, we also need to use autocast_context
# in caller to this method to ensure that it's called under the correct context
# (we've added `autocast_context` to `gen_track_with_mask_stream` in app.py).
with self.autocast_context(), self.inference_lock:
logger.info(
f"propagate in video in session {session_id}: "
f"{propagation_direction=}, {start_frame_idx=}, {max_frame_num_to_track=}"
)
try:
session = self.__get_session(session_id)
session["canceled"] = False
inference_state = session["state"]
if propagation_direction not in ["both", "forward", "backward"]:
raise ValueError(
f"invalid propagation direction: {propagation_direction}"
)
# First doing the forward propagation
if propagation_direction in ["both", "forward"]:
for outputs in self.predictor.propagate_in_video(
inference_state=inference_state,
start_frame_idx=start_frame_idx,
max_frame_num_to_track=max_frame_num_to_track,
reverse=False,
):
if session["canceled"]:
return None
frame_idx, obj_ids, video_res_masks = outputs
masks_binary = (
(video_res_masks > self.score_thresh)[:, 0].cpu().numpy()
)
rle_mask_list = self.__get_rle_mask_list(
object_ids=obj_ids, masks=masks_binary
)
yield PropagateDataResponse(
frame_index=frame_idx,
results=rle_mask_list,
)
# Then doing the backward propagation (reverse in time)
if propagation_direction in ["both", "backward"]:
for outputs in self.predictor.propagate_in_video(
inference_state=inference_state,
start_frame_idx=start_frame_idx,
max_frame_num_to_track=max_frame_num_to_track,
reverse=True,
):
if session["canceled"]:
return None
frame_idx, obj_ids, video_res_masks = outputs
masks_binary = (
(video_res_masks > self.score_thresh)[:, 0].cpu().numpy()
)
rle_mask_list = self.__get_rle_mask_list(
object_ids=obj_ids, masks=masks_binary
)
yield PropagateDataResponse(
frame_index=frame_idx,
results=rle_mask_list,
)
finally:
# Log upon completion (so that e.g. we can see if two propagations happen in parallel).
# Using `finally` here to log even when the tracking is aborted with GeneratorExit.
logger.info(
f"propagation ended in session {session_id}; {self.__get_session_stats()}"
)
def cancel_propagate_in_video(
self, request: CancelPropagateInVideoRequest
) -> CancelPorpagateResponse:
session = self.__get_session(request.session_id)
session["canceled"] = True
return CancelPorpagateResponse(success=True)
def __get_rle_mask_list(
self, object_ids: List[int], masks: np.ndarray
) -> List[PropagateDataValue]:
"""
Return a list of data values, i.e. list of object/mask combos.
"""
return [
self.__get_mask_for_object(object_id=object_id, mask=mask)
for object_id, mask in zip(object_ids, masks)
]
def __get_mask_for_object(
self, object_id: int, mask: np.ndarray
) -> PropagateDataValue:
"""
Create a data value for an object/mask combo.
"""
mask_rle = encode_masks(np.array(mask, dtype=np.uint8, order="F"))
mask_rle["counts"] = mask_rle["counts"].decode()
return PropagateDataValue(
object_id=object_id,
mask=Mask(
size=mask_rle["size"],
counts=mask_rle["counts"],
),
)
def __get_session(self, session_id: str):
session = self.session_states.get(session_id, None)
if session is None:
raise RuntimeError(
f"Cannot find session {session_id}; it might have expired"
)
return session
def __get_session_stats(self):
"""Get a statistics string for live sessions and their GPU usage."""
# print both the session ids and their video frame numbers
live_session_strs = [
f"'{session_id}' ({session['state']['num_frames']} frames, "
f"{len(session['state']['obj_ids'])} objects)"
for session_id, session in self.session_states.items()
]
session_stats_str = (
"Test String Here - -"
f"live sessions: [{', '.join(live_session_strs)}], GPU memory: "
f"{torch.cuda.memory_allocated() // 1024**2} MiB used and "
f"{torch.cuda.memory_reserved() // 1024**2} MiB reserved"
f" (max over time: {torch.cuda.max_memory_allocated() // 1024**2} MiB used "
f"and {torch.cuda.max_memory_reserved() // 1024**2} MiB reserved)"
)
return session_stats_str
def __clear_session_state(self, session_id: str) -> bool:
session = self.session_states.pop(session_id, None)
if session is None:
logger.warning(
f"cannot close session {session_id} as it does not exist (it might have expired); "
f"{self.__get_session_stats()}"
)
return False
else:
logger.info(f"removed session {session_id}; {self.__get_session_stats()}")
return True

BIN
sam2/demo/data/gallery/01_dog.mp4 Normal file
View File

Binary file not shown.

BIN
sam2/demo/data/gallery/02_cups.mp4 Normal file
View File

Binary file not shown.

BIN
sam2/demo/data/gallery/03_blocks.mp4 Normal file
View File

Binary file not shown.

BIN
sam2/demo/data/gallery/04_coffee.mp4 Normal file
View File

Binary file not shown.

BIN
sam2/demo/data/gallery/05_default_juggle.mp4 Normal file
View File

Binary file not shown.

7
sam2/demo/frontend/.babelrc Normal file
View File

@@ -0,0 +1,7 @@
{
"env": {
"production": {
"plugins": ["babel-plugin-strip-invariant"]
}
}
}

32
sam2/demo/frontend/.dockerignore Normal file
View File

@@ -0,0 +1,32 @@
# Logs
logs
*.log
npm-debug.log*
yarn-debug.log*
yarn-error.log*
pnpm-debug.log*
lerna-debug.log*
node_modules
dist
dist-ssr
*.local
storybook-static
.env
# Editor directories and files
.vscode/*
!.vscode/extensions.json
.idea
.DS_Store
*.suo
*.ntvs*
*.njsproj
*.sln
*.sw?
# Test results
/coverage/
/test-results/
/playwright-report/
/playwright/.cache/

3
sam2/demo/frontend/.eslintignore Normal file
View File

@@ -0,0 +1,3 @@
node_modules/
dist/
env.d.ts

63
sam2/demo/frontend/.eslintrc.cjs Normal file
View File

@@ -0,0 +1,63 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
module.exports = {
root: true,
env: {browser: true, es2020: true},
extends: [
'eslint:recommended',
'plugin:@typescript-eslint/recommended',
'plugin:react/recommended',
'plugin:react-hooks/recommended',
'plugin:import/recommended',
'plugin:prettier/recommended',
],
ignorePatterns: ['dist', '.eslintrc.cjs'],
parser: '@typescript-eslint/parser',
parserOptions: {
ecmaVersion: 'latest',
sourceType: 'module',
project: ['./tsconfig.json', './tsconfig.node.json'],
tsconfigRootDir: __dirname,
},
plugins: ['react-refresh'],
settings: {
react: {
version: 'detect',
},
'import/resolver': {
typescript: {},
node: {},
},
},
rules: {
'no-console': 'warn',
curly: 'warn',
'react/jsx-no-useless-fragment': 'warn',
'@typescript-eslint/no-unused-vars': [
'warn',
{
argsIgnorePattern: '^_',
},
],
'react-refresh/only-export-components': [
'warn',
{
allowConstantExport: true,
},
],
'react/react-in-jsx-scope': 'off',
},
};

32
sam2/demo/frontend/.gitignore vendored Normal file
View File

@@ -0,0 +1,32 @@
# Logs
logs
*.log
npm-debug.log*
yarn-debug.log*
yarn-error.log*
pnpm-debug.log*
lerna-debug.log*
node_modules
dist
dist-ssr
*.local
storybook-static
.env
# Editor directories and files
.vscode/*
!.vscode/extensions.json
.idea
.DS_Store
*.suo
*.ntvs*
*.njsproj
*.sln
*.sw?
# Test results
/coverage/
/test-results/
/playwright-report/
/playwright/.cache/

2
sam2/demo/frontend/.prettierignore Normal file
View File

@@ -0,0 +1,2 @@
node_modules/
dist/

9
sam2/demo/frontend/.prettierrc.json Normal file
View File

@@ -0,0 +1,9 @@
{
"arrowParens": "avoid",
"bracketSameLine": true,
"bracketSpacing": false,
"singleQuote": true,
"tabWidth": 2,
"trailingComma": "all",
"endOfLine": "auto"
}

1
sam2/demo/frontend/.watchmanconfig Normal file
View File

@@ -0,0 +1 @@
{}

26
sam2/demo/frontend/frontend.Dockerfile Normal file
View File

@@ -0,0 +1,26 @@
# Stage 1: Build Stage
FROM node:22.9.0 AS build
WORKDIR /app
# Copy package.json and yarn.lock
COPY package.json ./
COPY yarn.lock ./
# Install dependencies
RUN yarn install --frozen-lockfile
# Copy source code
COPY . .
# Build the application
RUN yarn build
# Stage 2: Production Stage
FROM nginx:latest
# Copy built files from the build stage to the production image
COPY --from=build /app/dist /usr/share/nginx/html
# Container startup command for the web server (nginx in this case)
CMD ["nginx", "-g", "daemon off;"]

14
sam2/demo/frontend/index.html Normal file
View File

@@ -0,0 +1,14 @@
<!doctype html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<meta
name="viewport"
content="width=device-width, initial-scale=1.0, shrink-to-fit=no" />
<title>SAM 2 Demo | By Meta FAIR</title>
</head>
<body>
<div id="root"></div>
<script type="module" src="/src/main.tsx"></script>
</body>
</html>

99
sam2/demo/frontend/package.json Normal file
View File

@@ -0,0 +1,99 @@
{
"name": "frontend-vite",
"private": true,
"version": "0.0.0",
"type": "module",
"scripts": {
"dev": "vite",
"merge-schemas": "tsx schemas/merge-schemas",
"relay": "yarn merge-schemas && relay-compiler",
"build": "tsc && vite build",
"lint": "eslint . --ext ts,tsx --report-unused-disable-directives --max-warnings 0",
"preview": "vite preview --open"
},
"dependencies": {
"@carbon/icons-react": "^11.34.1",
"@heroicons/react": "^2.0.18",
"@monaco-editor/react": "^4.6.0",
"@stylexjs/stylex": "^0.6.1",
"graphql": "^16.8.1",
"immer": "^10.0.3",
"immutability-helper": "^3.1.1",
"jotai": "^2.6.1",
"jotai-immer": "^0.3.0",
"localforage": "^1.10.0",
"monaco-editor": "^0.48.0",
"mp4box": "^0.5.2",
"pts": "^0.12.8",
"react": "^18.2.0",
"react-daisyui": "^4.1.0",
"react-device-detect": "^2.2.3",
"react-dom": "^18.2.0",
"react-dropzone": "^14.2.3",
"react-error-boundary": "^4.0.11",
"react-photo-album": "^2.3.0",
"react-pts-canvas": "^0.5.2",
"react-relay": "^16.2.0",
"react-router-dom": "^6.15.0",
"relay-runtime": "^16.2.0",
"serialize-error": "^11.0.3",
"use-immer": "^0.9.0",
"use-resize-observer": "^9.1.0"
},
"devDependencies": {
"@graphql-tools/load-files": "^7.0.0",
"@graphql-tools/merge": "^9.0.4",
"@tailwindcss/typography": "^0.5.9",
"@types/dom-webcodecs": "^0.1.11",
"@types/invariant": "^2.2.37",
"@types/node": "^20.14.10",
"@types/react": "^18.2.47",
"@types/react-dom": "^18.2.7",
"@types/react-relay": "^16.0.6",
"@types/relay-runtime": "^14.1.13",
"@typescript-eslint/eslint-plugin": "^6.18.1",
"@typescript-eslint/parser": "^6.18.1",
"@vitejs/plugin-react": "^4.2.1",
"autoprefixer": "^10.4.15",
"babel-plugin-relay": "^16.2.0",
"babel-plugin-strip-invariant": "^1.0.0",
"daisyui": "^3.6.3",
"eslint": "^8.48.0",
"eslint-config-prettier": "^9.0.0",
"eslint-import-resolver-alias": "^1.1.2",
"eslint-import-resolver-typescript": "^3.6.3",
"eslint-plugin-import": "^2.28.1",
"eslint-plugin-prettier": "^5.1.3",
"eslint-plugin-react": "^7.33.2",
"eslint-plugin-react-hooks": "^4.6.0",
"eslint-plugin-react-refresh": "^0.4.3",
"invariant": "^2.2.4",
"postcss": "^8.4.28",
"postinstall-postinstall": "^2.1.0",
"prettier": "^3.0.3",
"relay-compiler": "^16.2.0",
"sass": "^1.66.1",
"strip-ansi": "^7.1.0",
"tailwindcss": "^3.3.3",
"tsx": "^4.16.2",
"typescript": ">=4.3.5 <5.4.0",
"vite": "^5.0.11",
"vite-plugin-babel": "^1.2.0",
"vite-plugin-relay": "^2.0.0",
"vite-plugin-stylex-dev": "^0.5.2"
},
"resolutions": {
"wrap-ansi": "7.0.0"
},
"relay": {
"src": "./src/",
"schema": "./schema.graphql",
"language": "typescript",
"eagerEsModules": true,
"exclude": [
"**/node_modules/**",
"**/__mocks__/**",
"**/__generated__/**"
]
}
}

23
sam2/demo/frontend/postcss.config.js Normal file
View File

@@ -0,0 +1,23 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
export default {
plugins: {
'postcss-import': {},
'tailwindcss/nesting': {},
tailwindcss: {},
autoprefixer: {},
},
};

BIN
sam2/demo/frontend/public/fonts/Inter-VariableFont_opsz,wght.ttf Normal file
View File

Binary file not shown.

212
sam2/demo/frontend/schema.graphql Normal file
View File

@@ -0,0 +1,212 @@
input AddPointsInput {
sessionId: String!
frameIndex: Int!
clearOldPoints: Boolean!
objectId: Int!
labels: [Int!]!
points: [[Float!]!]!
}
type CancelPropagateInVideo {
success: Boolean!
}
input CancelPropagateInVideoInput {
sessionId: String!
}
input ClearPointsInFrameInput {
sessionId: String!
frameIndex: Int!
objectId: Int!
}
type ClearPointsInVideo {
success: Boolean!
}
input ClearPointsInVideoInput {
sessionId: String!
}
type CloseSession {
success: Boolean!
}
input CloseSessionInput {
sessionId: String!
}
type Mutation {
startSession(input: StartSessionInput!): StartSession!
closeSession(input: CloseSessionInput!): CloseSession!
addPoints(input: AddPointsInput!): RLEMaskListOnFrame!
clearPointsInFrame(input: ClearPointsInFrameInput!): RLEMaskListOnFrame!
clearPointsInVideo(input: ClearPointsInVideoInput!): ClearPointsInVideo!
removeObject(input: RemoveObjectInput!): [RLEMaskListOnFrame!]!
cancelPropagateInVideo(
input: CancelPropagateInVideoInput!
): CancelPropagateInVideo!
createDeletionId: String!
acceptTos: Boolean!
acceptTermsOfService: String!
uploadVideo(
file: Upload!
startTimeSec: Float = null
durationTimeSec: Float = null
): Video!
uploadSharedVideo(file: Upload!): SharedVideo!
uploadAnnotations(file: Upload!): Boolean!
}
input PingInput {
sessionId: String!
}
type Pong {
success: Boolean!
}
type Query {
ping(input: PingInput!): Pong!
defaultVideo: Video!
videos(
"""
Returns the items in the list that come before the specified cursor.
"""
before: String = null
"""
Returns the items in the list that come after the specified cursor.
"""
after: String = null
"""
Returns the first n items from the list.
"""
first: Int = null
"""
Returns the items in the list that come after the specified cursor.
"""
last: Int = null
): VideoConnection!
sharedVideo(path: String!): SharedVideo!
}
type RLEMask {
size: [Int!]!
counts: String!
order: String!
}
type RLEMaskForObject {
objectId: Int!
rleMask: RLEMask!
}
type RLEMaskListOnFrame {
frameIndex: Int!
rleMaskList: [RLEMaskForObject!]!
}
input RemoveObjectInput {
sessionId: String!
objectId: Int!
}
type StartSession {
sessionId: String!
}
input StartSessionInput {
path: String!
}
"""
The `ID` scalar type represents a unique identifier, often used to refetch an object or as key for a cache. The ID type appears in a JSON response as a String; however, it is not intended to be human-readable. When expected as an input type, any string (such as `"4"`) or integer (such as `4`) input value will be accepted as an ID.
"""
scalar GlobalID
@specifiedBy(url: "https://relay.dev/graphql/objectidentification.htm")
"""
An object with a Globally Unique ID
"""
interface Node {
"""
The Globally Unique ID of this object
"""
id: GlobalID!
}
"""
Information to aid in pagination.
"""
type PageInfo {
"""
When paginating forwards, are there more items?
"""
hasNextPage: Boolean!
"""
When paginating backwards, are there more items?
"""
hasPreviousPage: Boolean!
"""
When paginating backwards, the cursor to continue.
"""
startCursor: String
"""
When paginating forwards, the cursor to continue.
"""
endCursor: String
}
type SharedVideo {
path: String!
url: String!
}
scalar Upload
type Video implements Node {
"""
The Globally Unique ID of this object
"""
id: GlobalID!
path: String!
posterPath: String
width: Int!
height: Int!
url: String!
posterUrl: String!
}
"""
A connection to a list of items.
"""
type VideoConnection {
"""
Pagination data for this connection
"""
pageInfo: PageInfo!
"""
Contains the nodes in this connection
"""
edges: [VideoEdge!]!
}
"""
An edge in a connection.
"""
type VideoEdge {
"""
A cursor for use in pagination
"""
cursor: String!
"""
The item at the end of the edge
"""
node: Video!
}
schema {
query: Query
mutation: Mutation
}

105
sam2/demo/frontend/schemas/inference-api-schema.graphql Normal file
View File

@@ -0,0 +1,105 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
input AddPointsInput {
sessionId: String!
frameIndex: Int!
clearOldPoints: Boolean!
objectId: Int!
labels: [Int!]!
points: [[Float!]!]!
}
type CancelPropagateInVideo {
success: Boolean!
}
input CancelPropagateInVideoInput {
sessionId: String!
}
input ClearPointsInFrameInput {
sessionId: String!
frameIndex: Int!
objectId: Int!
}
type ClearPointsInVideo {
success: Boolean!
}
input ClearPointsInVideoInput {
sessionId: String!
}
type CloseSession {
success: Boolean!
}
input CloseSessionInput {
sessionId: String!
}
type Mutation {
startSession(input: StartSessionInput!): StartSession!
closeSession(input: CloseSessionInput!): CloseSession!
addPoints(input: AddPointsInput!): RLEMaskListOnFrame!
clearPointsInFrame(input: ClearPointsInFrameInput!): RLEMaskListOnFrame!
clearPointsInVideo(input: ClearPointsInVideoInput!): ClearPointsInVideo!
removeObject(input: RemoveObjectInput!): [RLEMaskListOnFrame!]!
cancelPropagateInVideo(
input: CancelPropagateInVideoInput!
): CancelPropagateInVideo!
}
input PingInput {
sessionId: String!
}
type Pong {
success: Boolean!
}
type Query {
ping(input: PingInput!): Pong!
}
type RLEMask {
size: [Int!]!
counts: String!
order: String!
}
type RLEMaskForObject {
objectId: Int!
rleMask: RLEMask!
}
type RLEMaskListOnFrame {
frameIndex: Int!
rleMaskList: [RLEMaskForObject!]!
}
input RemoveObjectInput {
sessionId: String!
objectId: Int!
}
type StartSession {
sessionId: String!
}
input StartSessionInput {
path: String!
}

33
sam2/demo/frontend/schemas/merge-schemas.ts Normal file
View File

@@ -0,0 +1,33 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {loadFilesSync} from '@graphql-tools/load-files';
import {mergeTypeDefs} from '@graphql-tools/merge';
import fs from 'fs';
import {print} from 'graphql';
import path from 'path';
import * as prettier from 'prettier';
import {fileURLToPath} from 'url';
const __filename = fileURLToPath(import.meta.url);
const __dirname = path.dirname(__filename);
const loadedFiles = loadFilesSync(`${__dirname}/*.graphql`);
const typeDefs = mergeTypeDefs(loadedFiles);
const printedTypeDefs = print(typeDefs);
const prettyTypeDefs = await prettier.format(printedTypeDefs, {
parser: 'graphql',
});
fs.writeFileSync('schema.graphql', prettyTypeDefs);

143
sam2/demo/frontend/schemas/video-api-schema.graphql Normal file
View File

@@ -0,0 +1,143 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
The `ID` scalar type represents a unique identifier, often used to refetch an object or as key for a cache. The ID type appears in a JSON response as a String; however, it is not intended to be human-readable. When expected as an input type, any string (such as `"4"`) or integer (such as `4`) input value will be accepted as an ID.
"""
scalar GlobalID
@specifiedBy(url: "https://relay.dev/graphql/objectidentification.htm")
type Mutation {
createDeletionId: String!
acceptTos: Boolean!
acceptTermsOfService: String!
uploadVideo(
file: Upload!
startTimeSec: Float = null
durationTimeSec: Float = null
): Video!
uploadSharedVideo(file: Upload!): SharedVideo!
uploadAnnotations(file: Upload!): Boolean!
}
"""
An object with a Globally Unique ID
"""
interface Node {
"""
The Globally Unique ID of this object
"""
id: GlobalID!
}
"""
Information to aid in pagination.
"""
type PageInfo {
"""
When paginating forwards, are there more items?
"""
hasNextPage: Boolean!
"""
When paginating backwards, are there more items?
"""
hasPreviousPage: Boolean!
"""
When paginating backwards, the cursor to continue.
"""
startCursor: String
"""
When paginating forwards, the cursor to continue.
"""
endCursor: String
}
type Query {
defaultVideo: Video!
videos(
"""
Returns the items in the list that come before the specified cursor.
"""
before: String = null
"""
Returns the items in the list that come after the specified cursor.
"""
after: String = null
"""
Returns the first n items from the list.
"""
first: Int = null
"""
Returns the items in the list that come after the specified cursor.
"""
last: Int = null
): VideoConnection!
sharedVideo(path: String!): SharedVideo!
}
type SharedVideo {
path: String!
url: String!
}
scalar Upload
type Video implements Node {
"""
The Globally Unique ID of this object
"""
id: GlobalID!
path: String!
posterPath: String
width: Int!
height: Int!
url: String!
posterUrl: String!
}
"""
A connection to a list of items.
"""
type VideoConnection {
"""
Pagination data for this connection
"""
pageInfo: PageInfo!
"""
Contains the nodes in this connection
"""
edges: [VideoEdge!]!
}
"""
An edge in a connection.
"""
type VideoEdge {
"""
A cursor for use in pagination
"""
cursor: String!
"""
The item at the end of the edge
"""
node: Video!
}

33
sam2/demo/frontend/src/App.tsx Normal file
View File

@@ -0,0 +1,33 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import SAM2DemoApp from '@/demo/SAM2DemoApp';
import SettingsContextProvider from '@/settings/SettingsContextProvider';
import {RouterProvider, createBrowserRouter} from 'react-router-dom';
export default function App() {
const router = createBrowserRouter([
{
path: '*',
element: (
<SettingsContextProvider>
<SAM2DemoApp />
</SettingsContextProvider>
),
},
]);
return <RouterProvider router={router} />;
}

BIN
sam2/demo/frontend/src/assets/icons/angery.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 16 KiB

BIN
sam2/demo/frontend/src/assets/icons/heart.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 19 KiB

BIN
sam2/demo/frontend/src/assets/icons/whistle.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 14 KiB

374
sam2/demo/frontend/src/assets/scss/App.scss Normal file
View File

@@ -0,0 +1,374 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
@tailwind base;
@tailwind components;
@tailwind utilities;
.tab {
display: flex;
padding: 0px 0px;
margin-right: 6px;
align-items: center;
height: 100%;
}
@layer base {
@font-face {
font-family: 'Inter';
src: url(/fonts/Inter-VariableFont.ttf) format('truetype-variations');
}
}
body {
font-family: 'Inter', sans-serif;
-webkit-font-smoothing: antialiased;
-moz-osx-font-smoothing: grayscale;
}
body,
html,
#root {
height: 100%;
@media screen and (max-width: '768px') {
overflow: hidden;
}
}
:root {
--segEv-font: 'Inter', system-ui, -apple-system, BlinkMacSystemFont,
'Segoe UI', Oxygen, Ubuntu, Cantarell, 'Open Sans', 'Helvetica Neue',
sans-serif;
--perspective: 4000px;
color-scheme: dark;
}
h1,
h2,
h3,
h4,
h5,
h6 {
font-family: 'Inter', sans-serif;
}
.prose .display h1 {
@apply text-4xl text-gray-800 font-medium leading-tight;
}
.prose .display h2 {
@apply text-gray-800 font-medium leading-tight;
font-size: 2.5rem;
}
.prose h1 {
@apply text-3xl text-gray-800 font-medium leading-tight mt-2 mb-4;
letter-spacing: 0.016rem;
}
.prose h2 {
@apply text-2xl text-gray-800 font-medium leading-tight my-2;
letter-spacing: 0.01rem;
}
.prose h3 {
@apply text-xl text-gray-800 font-medium leading-tight my-2;
letter-spacing: 0.005rem;
}
.prose h4 {
@apply text-lg text-gray-800 font-medium leading-tight my-2;
}
.prose h5 {
@apply text-xl text-gray-700 font-normal leading-normal my-2;
letter-spacing: 0.005rem;
}
.prose h6 {
@apply text-base text-gray-700 font-normal leading-normal my-2;
}
.prose p {
@apply text-sm text-gray-700 font-normal leading-normal;
@apply leading-snug;
}
.prose ol,
.prose ul {
@apply text-sm text-gray-700 font-normal leading-normal;
padding-right: 2rem;
}
.dark-mode h1,
.dark-mode h2,
.dark-mode h3,
.dark-mode h4,
.dark-mode h5,
.dark-mode p,
.dark-mode ol,
.dark-mode ul,
.dark-mode p *,
.dark-mode ol *,
.dark-mode ul *,
≈ {
@apply text-white;
}
.dark-mode h4,
.dark-mode h6,
.dark-mode li::marker,
.dark-mode a {
@apply text-gray-200;
}
.flex-grow-2 {
flex-grow: 2;
}
.flex-grow-3 {
flex-grow: 3;
}
.flex-grow-4 {
flex-grow: 4;
}
.flex-grow-5 {
flex-grow: 5;
}
.nav-title {
font-family: var(--segEv-font);
}
.segment-active {
animation: segment-highlight 2s linear infinite;
stroke-dasharray: 5, 10;
stroke-width: 4px;
}
@keyframes segment-highlight {
to {
stroke-dashoffset: 60;
}
}
.segment-select {
animation: segment-dotted 2s linear infinite;
stroke-dasharray: 3, 5;
stroke-width: 3px;
}
@keyframes segment-dotted {
to {
stroke-dashoffset: 24;
}
}
/**
* Daisy UI customizations
*/
.btn {
@apply normal-case rounded-md;
}
.comp_summary h1,
.comp_summary h2,
.comp_summary h3 {
@apply mb-4;
}
.disabled {
opacity: 0.4;
pointer-events: none;
}
.absolute-center {
top: 50%;
left: 50%;
transform: translate(-50%, -50%);
}
@screen lg {
.drawer .grid {
grid-template-columns: max-content 1fr;
}
}
.fade-in {
transition: opacity 0.5s;
opacity: 1 !important;
}
.react-photo-gallery--gallery > div {
gap: 0.25rem;
}
.sticker {
filter: drop-shadow(0.25rem 0.25rem 5px #fff)
drop-shadow(-0.25rem 0.25rem 5px #fff)
drop-shadow(0.25rem -0.25rem 5px #fff)
drop-shadow(-0.25rem -0.25rem 5px #fff);
transition: filter 0.3s ease-out;
}
.sticker:hover,
.sticker-select {
filter: drop-shadow(0.25rem 0.25rem 1px #2962d9)
drop-shadow(-0.25rem 0.25rem 1px #2962d9)
drop-shadow(0.25rem -0.25rem 1px #2962d9)
drop-shadow(-0.25rem -0.25rem 1px #2962d9);
}
/* keyframe animations */
.mask-path,
.reveal {
opacity: 0;
animation: reveal 0.4s ease-in forwards;
}
.slow-reveal {
animation: reveal 1s ease-in;
}
.reveal-then-conceal {
opacity: 0;
animation: reveal-then-conceal 1.5s ease-in-out forwards;
}
@keyframes reveal {
from {
opacity: 0;
}
to {
opacity: 1;
}
}
@keyframes reveal-then-conceal {
from {
opacity: 0;
}
50% {
opacity: 1;
}
to {
opacity: 0;
}
}
.background-animate {
background-size: 400%;
animation: pulse 3s ease infinite;
}
@keyframes pulse {
0%,
100% {
background-position: 0% 50%;
}
50% {
background-position: 100% 50%;
}
}
/* Fix for Safari and Mobile Safari:
Extracted Tailwind progress-bar styles and applied
them to a <div> instead of a <progress> element */
.loading-bar {
position: relative;
width: 100%;
-webkit-appearance: none;
-moz-appearance: none;
appearance: none;
overflow: hidden;
height: 0.5rem;
border-radius: 1rem;
border-radius: var(--rounded-box, 1rem);
vertical-align: baseline;
background-color: hsl(var(--n) / var(--tw-bg-opacity));
--tw-bg-opacity: 0.2;
&::after {
--tw-bg-opacity: 1;
background-color: hsl(var(--n) / var(--tw-bg-opacity));
content: '';
position: absolute;
top: 0px;
bottom: 0px;
left: -40%;
width: 33.333333%;
border-radius: 1rem;
border-radius: var(--rounded-box, 1rem);
animation: loading 5s infinite ease-in-out;
}
}
@keyframes loading {
50% {
left: 107%;
}
}
@keyframes inAnimation {
0% {
opacity: 0;
max-height: 0px;
}
50% {
opacity: 1;
}
100% {
opacity: 1;
max-height: 600px;
}
}
@keyframes outAnimation {
0% {
opacity: 1;
max-height: 600px;
}
50% {
opacity: 0;
}
100% {
opacity: 0;
max-height: 0px;
}
}
@keyframes ellipsisAnimation {
0% {
content: '';
}
25% {
content: '.';
}
50% {
content: '..';
}
75% {
content: '...';
}
}
.ellipsis::after {
content: '';
animation: ellipsisAnimation 1.5s infinite;
}

BIN
sam2/demo/frontend/src/assets/videos/sam2_720px_dark.mp4 Normal file
View File

Binary file not shown.

BIN
sam2/demo/frontend/src/assets/videos/sam2_video_poster.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 8.6 KiB

284
sam2/demo/frontend/src/common/codecs/VideoDecoder.ts Normal file
View File

@@ -0,0 +1,284 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {cloneFrame} from '@/common/codecs/WebCodecUtils';
import {FileStream} from '@/common/utils/FileUtils';
import {
createFile,
DataStream,
MP4ArrayBuffer,
MP4File,
MP4Sample,
MP4VideoTrack,
} from 'mp4box';
import {isAndroid, isChrome, isEdge, isWindows} from 'react-device-detect';
export type ImageFrame = {
bitmap: VideoFrame;
timestamp: number;
duration: number;
};
export type DecodedVideo = {
width: number;
height: number;
frames: ImageFrame[];
numFrames: number;
fps: number;
};
function decodeInternal(
identifier: string,
onReady: (mp4File: MP4File) => Promise<void>,
onProgress: (decodedVideo: DecodedVideo) => void,
): Promise<DecodedVideo> {
return new Promise((resolve, reject) => {
const imageFrames: ImageFrame[] = [];
const globalSamples: MP4Sample[] = [];
let decoder: VideoDecoder;
let track: MP4VideoTrack | null = null;
const mp4File = createFile();
mp4File.onError = reject;
mp4File.onReady = async info => {
if (info.videoTracks.length > 0) {
track = info.videoTracks[0];
} else {
// The video does not have a video track, so looking if there is an
// "otherTracks" available. Note, I couldn't find any documentation
// about "otherTracks" in WebCodecs [1], but it was available in the
// info for MP4V-ES, which isn't supported by Chrome [2].
// However, we'll still try to get the track and then throw an error
// further down in the VideoDecoder.isConfigSupported if the codec is
// not supported by the browser.
//
// [1] https://www.w3.org/TR/webcodecs/
// [2] https://developer.mozilla.org/en-US/docs/Web/Media/Formats/Video_codecs#mp4v-es
track = info.otherTracks[0];
}
if (track == null) {
reject(new Error(`${identifier} does not contain a video track`));
return;
}
const timescale = track.timescale;
const edits = track.edits;
let frame_n = 0;
decoder = new VideoDecoder({
// Be careful with any await in this function. The VideoDecoder will
// not await output and continue calling it with decoded frames.
async output(inputFrame) {
if (track == null) {
reject(new Error(`${identifier} does not contain a video track`));
return;
}
const saveTrack = track;
// If the track has edits, we'll need to check that only frames are
// returned that are within the edit list. This can happen for
// trimmed videos that have not been transcoded and therefore the
// video track contains more frames than those visually rendered when
// playing back the video.
if (edits != null && edits.length > 0) {
const cts = Math.round(
(inputFrame.timestamp * timescale) / 1_000_000,
);
if (cts < edits[0].media_time) {
inputFrame.close();
return;
}
}
// Workaround for Chrome where the decoding stops at ~17 frames unless
// the VideoFrame is closed. So, the workaround here is to create a
// new VideoFrame and close the decoded VideoFrame.
// The frame has to be cloned, or otherwise some frames at the end of the
// video will be black. Note, the default VideoFrame.clone doesn't work
// and it is using a frame cloning found here:
// https://webcodecs-blogpost-demo.glitch.me/
if (
(isAndroid && isChrome) ||
(isWindows && isChrome) ||
(isWindows && isEdge)
) {
const clonedFrame = await cloneFrame(inputFrame);
inputFrame.close();
inputFrame = clonedFrame;
}
const sample = globalSamples[frame_n];
if (sample != null) {
const duration = (sample.duration * 1_000_000) / sample.timescale;
imageFrames.push({
bitmap: inputFrame,
timestamp: inputFrame.timestamp,
duration,
});
// Sort frames in order of timestamp. This is needed because Safari
// can return decoded frames out of order.
imageFrames.sort((a, b) => (a.timestamp < b.timestamp ? 1 : -1));
// Update progress on first frame and then every 40th frame
if (onProgress != null && frame_n % 100 === 0) {
onProgress({
width: saveTrack.track_width,
height: saveTrack.track_height,
frames: imageFrames,
numFrames: saveTrack.nb_samples,
fps:
(saveTrack.nb_samples / saveTrack.duration) *
saveTrack.timescale,
});
}
}
frame_n++;
if (saveTrack.nb_samples === frame_n) {
// Sort frames in order of timestamp. This is needed because Safari
// can return decoded frames out of order.
imageFrames.sort((a, b) => (a.timestamp > b.timestamp ? 1 : -1));
resolve({
width: saveTrack.track_width,
height: saveTrack.track_height,
frames: imageFrames,
numFrames: saveTrack.nb_samples,
fps:
(saveTrack.nb_samples / saveTrack.duration) *
saveTrack.timescale,
});
}
},
error(error) {
reject(error);
},
});
let description;
const trak = mp4File.getTrackById(track.id);
const entries = trak?.mdia?.minf?.stbl?.stsd?.entries;
if (entries == null) {
return;
}
for (const entry of entries) {
if (entry.avcC || entry.hvcC) {
const stream = new DataStream(undefined, 0, DataStream.BIG_ENDIAN);
if (entry.avcC) {
entry.avcC.write(stream);
} else if (entry.hvcC) {
entry.hvcC.write(stream);
}
description = new Uint8Array(stream.buffer, 8); // Remove the box header.
break;
}
}
const configuration: VideoDecoderConfig = {
codec: track.codec,
codedWidth: track.track_width,
codedHeight: track.track_height,
description,
};
const supportedConfig =
await VideoDecoder.isConfigSupported(configuration);
if (supportedConfig.supported == true) {
decoder.configure(configuration);
mp4File.setExtractionOptions(track.id, null, {
nbSamples: Infinity,
});
mp4File.start();
} else {
reject(
new Error(
`Decoder config faile: config ${JSON.stringify(
supportedConfig.config,
)} is not supported`,
),
);
return;
}
};
mp4File.onSamples = async (
_id: number,
_user: unknown,
samples: MP4Sample[],
) => {
for (const sample of samples) {
globalSamples.push(sample);
decoder.decode(
new EncodedVideoChunk({
type: sample.is_sync ? 'key' : 'delta',
timestamp: (sample.cts * 1_000_000) / sample.timescale,
duration: (sample.duration * 1_000_000) / sample.timescale,
data: sample.data,
}),
);
}
await decoder.flush();
decoder.close();
};
onReady(mp4File);
});
}
export function decode(
file: File,
onProgress: (decodedVideo: DecodedVideo) => void,
): Promise<DecodedVideo> {
return decodeInternal(
file.name,
async (mp4File: MP4File) => {
const reader = new FileReader();
reader.onload = function () {
const result = this.result as MP4ArrayBuffer;
if (result != null) {
result.fileStart = 0;
mp4File.appendBuffer(result);
}
mp4File.flush();
};
reader.readAsArrayBuffer(file);
},
onProgress,
);
}
export function decodeStream(
fileStream: FileStream,
onProgress: (decodedVideo: DecodedVideo) => void,
): Promise<DecodedVideo> {
return decodeInternal(
'stream',
async (mp4File: MP4File) => {
let part = await fileStream.next();
while (part.done === false) {
const result = part.value.data.buffer as MP4ArrayBuffer;
if (result != null) {
result.fileStart = part.value.range.start;
mp4File.appendBuffer(result);
}
mp4File.flush();
part = await fileStream.next();
}
},
onProgress,
);
}

139
sam2/demo/frontend/src/common/codecs/VideoEncoder.ts Normal file
View File

@@ -0,0 +1,139 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {ImageFrame} from '@/common/codecs/VideoDecoder';
import {MP4ArrayBuffer, createFile} from 'mp4box';
// The selection of timescale and seconds/key-frame value are
// explained in the following docs: https://github.com/vjeux/mp4-h264-re-encode
const TIMESCALE = 90000;
const SECONDS_PER_KEY_FRAME = 2;
export function encode(
width: number,
height: number,
numFrames: number,
framesGenerator: AsyncGenerator<ImageFrame, unknown>,
progressCallback?: (progress: number) => void,
): Promise<MP4ArrayBuffer> {
return new Promise((resolve, reject) => {
let encodedFrameIndex = 0;
let nextKeyFrameTimestamp = 0;
let trackID: number | null = null;
const durations: number[] = [];
const outputFile = createFile();
const encoder = new VideoEncoder({
output(chunk, metaData) {
const uint8 = new Uint8Array(chunk.byteLength);
chunk.copyTo(uint8);
const description = metaData?.decoderConfig?.description;
if (trackID === null) {
trackID = outputFile.addTrack({
width: width,
height: height,
timescale: TIMESCALE,
avcDecoderConfigRecord: description,
});
}
const shiftedDuration = durations.shift();
if (shiftedDuration != null) {
outputFile.addSample(trackID, uint8, {
duration: getScaledDuration(shiftedDuration),
is_sync: chunk.type === 'key',
});
encodedFrameIndex++;
progressCallback?.(encodedFrameIndex / numFrames);
}
if (encodedFrameIndex === numFrames) {
resolve(outputFile.getBuffer());
}
},
error(error) {
reject(error);
return;
},
});
const setConfigurationAndEncodeFrames = async () => {
// The codec value was taken from the following implementation and seems
// reasonable for our use case for now:
// https://github.com/vjeux/mp4-h264-re-encode/blob/main/mp4box.html#L103
// Additional details about codecs can be found here:
// - https://developer.mozilla.org/en-US/docs/Web/Media/Formats/codecs_parameter
// - https://www.w3.org/TR/webcodecs-codec-registry/#video-codec-registry
//
// The following setting is a good compromise between output video file
// size and quality. The latencyMode "realtime" is needed for Safari,
// which otherwise will produce 20x larger files when in quality
// latencyMode. Chrome does a really good job with file size even when
// latencyMode is set to quality.
const configuration: VideoEncoderConfig = {
codec: 'avc1.4d0034',
width: roundToNearestEven(width),
height: roundToNearestEven(height),
bitrate: 14_000_000,
alpha: 'discard',
bitrateMode: 'variable',
latencyMode: 'realtime',
};
const supportedConfig =
await VideoEncoder.isConfigSupported(configuration);
if (supportedConfig.supported === true) {
encoder.configure(configuration);
} else {
throw new Error(
`Unsupported video encoder config ${JSON.stringify(supportedConfig)}`,
);
}
for await (const frame of framesGenerator) {
const {bitmap, duration, timestamp} = frame;
durations.push(duration);
let keyFrame = false;
if (timestamp >= nextKeyFrameTimestamp) {
await encoder.flush();
keyFrame = true;
nextKeyFrameTimestamp = timestamp + SECONDS_PER_KEY_FRAME * 1e6;
}
encoder.encode(bitmap, {keyFrame});
bitmap.close();
}
await encoder.flush();
encoder.close();
};
setConfigurationAndEncodeFrames();
});
}
function getScaledDuration(rawDuration: number) {
return rawDuration / (1_000_000 / TIMESCALE);
}
function roundToNearestEven(dim: number) {
const rounded = Math.round(dim);
if (rounded % 2 === 0) {
return rounded;
} else {
return rounded + (rounded > dim ? -1 : 1);
}
}

50
sam2/demo/frontend/src/common/codecs/WebCodecUtils.ts Normal file
View File

@@ -0,0 +1,50 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
// https://github.com/w3c/webcodecs/issues/88
// https://issues.chromium.org/issues/40725065
// https://webcodecs-blogpost-demo.glitch.me/
export async function cloneFrame(frame: VideoFrame): Promise<VideoFrame> {
const {
codedHeight,
codedWidth,
colorSpace,
displayHeight,
displayWidth,
format,
timestamp,
} = frame;
const rect = {x: 0, y: 0, width: codedWidth, height: codedHeight};
const data = new ArrayBuffer(frame.allocationSize({rect}));
try {
await frame.copyTo(data, {rect});
} catch (error) {
// The VideoFrame#copyTo on x64 builds on macOS fails. The workaround here
// is to clone the frame.
// https://stackoverflow.com/questions/77898766/inconsistent-behavior-of-webcodecs-copyto-method-across-different-browsers-an
return frame.clone();
}
return new VideoFrame(data, {
codedHeight,
codedWidth,
colorSpace,
displayHeight,
displayWidth,
duration: frame.duration ?? undefined,
format: format!,
timestamp,
visibleRect: frame.visibleRect ?? undefined,
});
}

74
sam2/demo/frontend/src/common/components/MobileFirstClickBanner.tsx Normal file
View File

@@ -0,0 +1,74 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ChangeVideoModal from '@/common/components/gallery/ChangeVideoModal';
import {DEMO_SHORT_NAME} from '@/demo/DemoConfig';
import {spacing} from '@/theme/tokens.stylex';
import {ImageCopy} from '@carbon/icons-react';
import stylex from '@stylexjs/stylex';
import {Button} from 'react-daisyui';
const styles = stylex.create({
container: {
position: 'relative',
backgroundColor: '#000',
padding: spacing[5],
paddingVertical: spacing[6],
display: 'flex',
flexDirection: 'column',
gap: spacing[4],
},
});
export default function MobileFirstClickBanner() {
return (
<div {...stylex.props(styles.container)}>
<div className="flex text-white text-lg">
Click an object in the video to start
</div>
<div className="text-sm text-[#A7B3BF]">
<p>
You&apos;ll be able to use {DEMO_SHORT_NAME} to make fun edits to any
video by tracking objects and applying visual effects. To start, click
any object in the video.
</p>
</div>
<div className="flex items-center">
<ChangeVideoModal
videoGalleryModalTrigger={MobileVideoGalleryModalTrigger}
showUploadInGallery={true}
/>
</div>
</div>
);
}
type MobileVideoGalleryModalTriggerProps = {
onClick: () => void;
};
function MobileVideoGalleryModalTrigger({
onClick,
}: MobileVideoGalleryModalTriggerProps) {
return (
<Button
color="ghost"
startIcon={<ImageCopy size={20} />}
onClick={onClick}
className="text-white p-0">
Change video
</Button>
);
}

41
sam2/demo/frontend/src/common/components/Tooltip.tsx Normal file
View File

@@ -0,0 +1,41 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {PropsWithChildren} from 'react';
type Props = PropsWithChildren<{
className?: string;
message: string;
position?: 'left' | 'top' | 'right' | 'bottom';
}>;
/**
* This is a custom Tooltip component because React Daisy UI does not have an
* option to *only* show tooltip on large devices.
*/
export default function Tooltip({
children,
className = '',
message,
position = 'top',
}: Props) {
return (
<div
className={`lg:tooltip tooltip-${position} ${className}`}
data-tip={message}>
{children}
</div>
);
}

49
sam2/demo/frontend/src/common/components/annotations/AddObjectButton.tsx Normal file
View File

@@ -0,0 +1,49 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import useMessagesSnackbar from '@/common/components/snackbar/useDemoMessagesSnackbar';
import useVideo from '@/common/components/video/editor/useVideo';
import {activeTrackletObjectIdAtom, labelTypeAtom} from '@/demo/atoms';
import {Add} from '@carbon/icons-react';
import {useSetAtom} from 'jotai';
export default function AddObjectButton() {
const video = useVideo();
const setActiveTrackletId = useSetAtom(activeTrackletObjectIdAtom);
const setLabelType = useSetAtom(labelTypeAtom);
const {enqueueMessage} = useMessagesSnackbar();
async function addObject() {
enqueueMessage('addObjectClick');
const tracklet = await video?.createTracklet();
if (tracklet != null) {
setActiveTrackletId(tracklet.id);
setLabelType('positive');
}
}
return (
<div
onClick={addObject}
className="group flex justify-start mx-4 px-4 bg-transparent text-white !rounded-xl border-none cursor-pointer">
<div className="flex gap-6 items-center">
<div className=" group-hover:bg-graydark-700 border border-white relative h-12 w-12 md:w-20 md:h-20 shrink-0 rounded-lg flex items-center justify-center">
<Add size={36} className="group-hover:text-white text-gray-300" />
</div>
<div className="font-medium text-base">Add another object</div>
</div>
</div>
);
}

81
sam2/demo/frontend/src/common/components/annotations/ClearAllPointsInVideoButton.tsx Normal file
View File

@@ -0,0 +1,81 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import useRestartSession from '@/common/components/session/useRestartSession';
import useMessagesSnackbar from '@/common/components/snackbar/useDemoMessagesSnackbar';
import useVideo from '@/common/components/video/editor/useVideo';
import {isPlayingAtom, isStreamingAtom, labelTypeAtom} from '@/demo/atoms';
import {Reset} from '@carbon/icons-react';
import stylex from '@stylexjs/stylex';
import {useAtomValue, useSetAtom} from 'jotai';
import {useState} from 'react';
import {Button, Loading} from 'react-daisyui';
const styles = stylex.create({
container: {
display: 'flex',
alignItems: 'center',
},
});
type Props = {
onRestart: () => void;
};
export default function ClearAllPointsInVideoButton({onRestart}: Props) {
const [isLoading, setIsLoading] = useState<boolean>(false);
const isPlaying = useAtomValue(isPlayingAtom);
const isStreaming = useAtomValue(isStreamingAtom);
const setLabelType = useSetAtom(labelTypeAtom);
const {clearMessage} = useMessagesSnackbar();
const {restartSession} = useRestartSession();
const video = useVideo();
async function handleRestart() {
if (video === null) {
return;
}
setIsLoading(true);
if (isPlaying) {
video.pause();
}
if (isStreaming) {
await video.abortStreamMasks();
}
const isSuccessful = await video.clearPointsInVideo();
if (!isSuccessful) {
await restartSession();
}
video.frame = 0;
setLabelType('positive');
onRestart();
clearMessage();
setIsLoading(false);
}
return (
<div {...stylex.props(styles.container)}>
<Button
color="ghost"
onClick={handleRestart}
className="!px-4 !rounded-full font-medium text-white hover:bg-black"
startIcon={isLoading ? <Loading size="sm" /> : <Reset size={20} />}>
Start over
</Button>
</div>
);
}

38
sam2/demo/frontend/src/common/components/annotations/CloseSessionButton.tsx Normal file
View File

@@ -0,0 +1,38 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import PrimaryCTAButton from '@/common/components/button/PrimaryCTAButton';
import useVideo from '@/common/components/video/editor/useVideo';
import {ChevronRight} from '@carbon/icons-react';
type Props = {
onSessionClose: () => void;
};
export default function CloseSessionButton({onSessionClose}: Props) {
const video = useVideo();
function handleCloseSession() {
video?.closeSession();
video?.logAnnotations();
onSessionClose();
}
return (
<PrimaryCTAButton onClick={handleCloseSession} endIcon={<ChevronRight />}>
Good to go
</PrimaryCTAButton>
);
}

49
sam2/demo/frontend/src/common/components/annotations/FirstClickView.tsx Normal file
View File

@@ -0,0 +1,49 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ChangeVideo from '@/common/components/gallery/ChangeVideoModal';
import useMessagesSnackbar from '@/common/components/snackbar/useDemoMessagesSnackbar';
import {DEMO_SHORT_NAME} from '@/demo/DemoConfig';
import {useEffect, useRef} from 'react';
export default function FirstClickView() {
const isFirstClickMessageShown = useRef(false);
const {enqueueMessage} = useMessagesSnackbar();
useEffect(() => {
if (!isFirstClickMessageShown.current) {
isFirstClickMessageShown.current = true;
enqueueMessage('firstClick');
}
}, [enqueueMessage]);
return (
<div className="w-full h-full flex flex-col p-8">
<div className="grow flex flex-col gap-6">
<h2 className="text-2xl">Click an object in the video to start</h2>
<p className="!text-gray-60">
You&apos;ll be able to use {DEMO_SHORT_NAME} to make fun edits to any
video by tracking objects and applying visual effects.
</p>
<p className="!text-gray-60">
To start, click any object in the video.
</p>
</div>
<div className="flex items-center">
<ChangeVideo />
</div>
</div>
);
}

30
sam2/demo/frontend/src/common/components/annotations/LimitNotice.tsx Normal file
View File

@@ -0,0 +1,30 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {InformationFilled} from '@carbon/icons-react';
export default function LimitNotice() {
return (
<div className="mt-6 gap-3 mx-6 flex items-center text-gray-400">
<div>
<InformationFilled size={32} />
</div>
<div className="text-sm leading-snug">
In this demo, you can track up to 3 objects, even though the SAM 2 model
does not have a limit.
</div>
</div>
);
}

78
sam2/demo/frontend/src/common/components/annotations/MobileObjectsList.tsx Normal file
View File

@@ -0,0 +1,78 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ClearAllPointsInVideoButton from '@/common/components/annotations/ClearAllPointsInVideoButton';
import ObjectThumbnail from '@/common/components/annotations/ObjectThumbnail';
import {OBJECT_TOOLBAR_INDEX} from '@/common/components/toolbar/ToolbarConfig';
import {BaseTracklet} from '@/common/tracker/Tracker';
import {activeTrackletObjectIdAtom, trackletObjectsAtom} from '@/demo/atoms';
import {spacing} from '@/theme/tokens.stylex';
import stylex from '@stylexjs/stylex';
import {useAtomValue, useSetAtom} from 'jotai';
const styles = stylex.create({
container: {
display: 'flex',
padding: spacing[5],
borderTop: '1px solid #DEE3E9',
},
trackletsContainer: {
flexGrow: 1,
display: 'flex',
alignItems: 'center',
gap: spacing[5],
},
});
type Props = {
showActiveObject: () => void;
onTabChange: (newIndex: number) => void;
};
export default function MobileObjectsList({
showActiveObject,
onTabChange,
}: Props) {
const tracklets = useAtomValue(trackletObjectsAtom);
const setActiveTrackletId = useSetAtom(activeTrackletObjectIdAtom);
function handleSelectObject(tracklet: BaseTracklet) {
setActiveTrackletId(tracklet.id);
showActiveObject();
}
return (
<div {...stylex.props(styles.container)}>
<div {...stylex.props(styles.trackletsContainer)}>
{tracklets.map(tracklet => {
const {id, color, thumbnail} = tracklet;
return (
<ObjectThumbnail
key={id}
color={color}
thumbnail={thumbnail}
onClick={() => {
handleSelectObject(tracklet);
}}
/>
);
})}
</div>
<ClearAllPointsInVideoButton
onRestart={() => onTabChange(OBJECT_TOOLBAR_INDEX)}
/>
</div>
);
}

51
sam2/demo/frontend/src/common/components/annotations/MobileObjectsToolbar.tsx Normal file
View File

@@ -0,0 +1,51 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import MobileObjectsToolbarHeader from '@/common/components/annotations/MobileObjectsToolbarHeader';
import ObjectsToolbarBottomActions from '@/common/components/annotations/ObjectsToolbarBottomActions';
import {getObjectLabel} from '@/common/components/annotations/ObjectUtils';
import ToolbarObject from '@/common/components/annotations/ToolbarObject';
import MobileFirstClickBanner from '@/common/components/MobileFirstClickBanner';
import {activeTrackletObjectAtom, isFirstClickMadeAtom} from '@/demo/atoms';
import {useAtomValue} from 'jotai';
type Props = {
onTabChange: (newIndex: number) => void;
};
export default function MobileObjectsToolbar({onTabChange}: Props) {
const activeTracklet = useAtomValue(activeTrackletObjectAtom);
const isFirstClickMade = useAtomValue(isFirstClickMadeAtom);
if (!isFirstClickMade) {
return <MobileFirstClickBanner />;
}
return (
<div className="w-full">
<MobileObjectsToolbarHeader />
{activeTracklet != null && (
<ToolbarObject
label={getObjectLabel(activeTracklet)}
tracklet={activeTracklet}
isActive={true}
isMobile={true}
/>
)}
<ObjectsToolbarBottomActions onTabChange={onTabChange} />
</div>
);
}

36
sam2/demo/frontend/src/common/components/annotations/MobileObjectsToolbarHeader.tsx Normal file
View File

@@ -0,0 +1,36 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ToolbarProgressChip from '@/common/components/toolbar/ToolbarProgressChip';
import {isStreamingAtom, streamingStateAtom} from '@/demo/atoms';
import {useAtomValue} from 'jotai';
export default function MobileObjectsToolbarHeader() {
const isStreaming = useAtomValue(isStreamingAtom);
const streamingState = useAtomValue(streamingStateAtom);
return (
<div className="w-full flex gap-4 items-center px-5 py-5">
<div className="grow text-sm text-white">
<ToolbarProgressChip />
{streamingState === 'full'
? 'Review your selected objects across the video, and continue to edit if needed. Once everything looks good, press “Next” to continue.'
: isStreaming
? 'Watch the video closely for any places where your objects aren’t tracked correctly. You can also stop tracking to make additional edits.'
: 'Edit your object selection with a few more clicks if needed. Press “Track objects” to track your objects throughout the video.'}
</div>
</div>
);
}

116
sam2/demo/frontend/src/common/components/annotations/ObjectActions.tsx Normal file
View File

@@ -0,0 +1,116 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import PointsToggle from '@/common/components/annotations/PointsToggle';
import useVideo from '@/common/components/video/editor/useVideo';
import useReportError from '@/common/error/useReportError';
import {
activeTrackletObjectIdAtom,
isPlayingAtom,
isStreamingAtom,
} from '@/demo/atoms';
import {
AddFilled,
Select_02,
SubtractFilled,
TrashCan,
} from '@carbon/icons-react';
import {useAtom, useAtomValue} from 'jotai';
import {useState} from 'react';
import type {ButtonProps} from 'react-daisyui';
import {Button} from 'react-daisyui';
type Props = {
objectId: number;
active: boolean;
};
function CustomButton({className, ...props}: ButtonProps) {
return (
<Button
size="sm"
color="ghost"
className={`font-medium border-none hover:bg-black px-2 h-10 ${className}`}
{...props}>
{props.children}
</Button>
);
}
export default function ObjectActions({objectId, active}: Props) {
const [isRemovingObject, setIsRemovingObject] = useState<boolean>(false);
const [activeTrackId, setActiveTrackletId] = useAtom(
activeTrackletObjectIdAtom,
);
const isStreaming = useAtomValue(isStreamingAtom);
const isPlaying = useAtom(isPlayingAtom);
const video = useVideo();
const reportError = useReportError();
async function handleRemoveObject(
event: React.MouseEvent<HTMLButtonElement>,
) {
try {
event.stopPropagation();
setIsRemovingObject(true);
if (isStreaming) {
await video?.abortStreamMasks();
}
if (isPlaying) {
video?.pause();
}
await video?.deleteTracklet(objectId);
} catch (error) {
reportError(error);
} finally {
setIsRemovingObject(false);
if (activeTrackId === objectId) {
setActiveTrackletId(null);
}
}
}
return (
<div>
{active && (
<div className="text-sm mt-1 leading-snug text-gray-400 hidden md:block ml-2 md:mb-4">
Select <AddFilled size={14} className="inline" /> to add areas to the
object and <SubtractFilled size={14} className="inline" /> to remove
areas from the object in the video. Click on an existing point to
delete it.
</div>
)}
<div className="flex justify-between items-center md:mt-2 mt-0">
{active ? (
<PointsToggle />
) : (
<>
<CustomButton startIcon={<Select_02 size={24} />}>
Edit selection
</CustomButton>
<CustomButton
loading={isRemovingObject}
onClick={handleRemoveObject}
startIcon={!isRemovingObject && <TrashCan size={24} />}>
<span className="hidden md:inline">Clear</span>
</CustomButton>
</>
)}
</div>
</div>
);
}

46
sam2/demo/frontend/src/common/components/annotations/ObjectPlaceholder.tsx Normal file
View File

@@ -0,0 +1,46 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {BLUE_PINK_FILL_BR} from '@/theme/gradientStyle';
type Props = {
showPlus?: boolean;
onClick?: () => void;
};
export default function ObjectPlaceholder({showPlus = true, onClick}: Props) {
return (
<div
className={`relative ${BLUE_PINK_FILL_BR} h-12 w-12 md:h-20 md:w-20 shrink-0 rounded-lg`}
onClick={onClick}>
{showPlus && (
<div className="absolute left-1/2 top-1/2 -translate-x-1/2 -translate-y-1/2">
<svg
xmlns="http://www.w3.org/2000/svg"
width="16"
height="16"
viewBox="0 0 16 16"
fill="none">
<path
d="M16 7H9V0H7V7H0V9H7V16H9V9H16V7Z"
fill="#667788"
fillOpacity={1}
/>
</svg>
</div>
)}
</div>
);
}

37
sam2/demo/frontend/src/common/components/annotations/ObjectThumbnail.tsx Normal file
View File

@@ -0,0 +1,37 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
type Props = {
thumbnail: string | null;
color: string;
onClick?: () => void;
};
export default function ObjectThumbnail({thumbnail, color, onClick}: Props) {
return (
<div
className="relative h-12 w-12 md:w-20 md:h-20 shrink-0 p-2 rounded-lg bg-contain bg-no-repeat bg-center"
style={{
backgroundColor: color,
}}
onClick={onClick}>
<div
className="w-full h-full bg-contain bg-no-repeat bg-center"
style={{
backgroundImage: thumbnail == null ? 'none' : `url(${thumbnail})`,
}}></div>
</div>
);
}

20
sam2/demo/frontend/src/common/components/annotations/ObjectUtils.ts Normal file
View File

@@ -0,0 +1,20 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {BaseTracklet} from '@/common/tracker/Tracker';
export function getObjectLabel(tracklet: BaseTracklet) {
return `Object ${tracklet.id + 1}`;
}

72
sam2/demo/frontend/src/common/components/annotations/ObjectsToolbar.tsx Normal file
View File

@@ -0,0 +1,72 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import AddObjectButton from '@/common/components/annotations/AddObjectButton';
import FirstClickView from '@/common/components/annotations/FirstClickView';
import LimitNotice from '@/common/components/annotations/LimitNotice';
import ObjectsToolbarBottomActions from '@/common/components/annotations/ObjectsToolbarBottomActions';
import ObjectsToolbarHeader from '@/common/components/annotations/ObjectsToolbarHeader';
import {getObjectLabel} from '@/common/components/annotations/ObjectUtils';
import ToolbarObject from '@/common/components/annotations/ToolbarObject';
import {
activeTrackletObjectAtom,
activeTrackletObjectIdAtom,
isAddObjectEnabledAtom,
isFirstClickMadeAtom,
isTrackletObjectLimitReachedAtom,
trackletObjectsAtom,
} from '@/demo/atoms';
import {useAtomValue, useSetAtom} from 'jotai';
type Props = {
onTabChange: (newIndex: number) => void;
};
export default function ObjectsToolbar({onTabChange}: Props) {
const tracklets = useAtomValue(trackletObjectsAtom);
const activeTracklet = useAtomValue(activeTrackletObjectAtom);
const setActiveTrackletId = useSetAtom(activeTrackletObjectIdAtom);
const isFirstClickMade = useAtomValue(isFirstClickMadeAtom);
const isObjectLimitReached = useAtomValue(isTrackletObjectLimitReachedAtom);
const isAddObjectEnabled = useAtomValue(isAddObjectEnabledAtom);
if (!isFirstClickMade) {
return <FirstClickView />;
}
return (
<div className="flex flex-col h-full">
<ObjectsToolbarHeader />
<div className="grow w-full overflow-y-auto">
{tracklets.map(tracklet => {
return (
<ToolbarObject
key={tracklet.id}
label={getObjectLabel(tracklet)}
tracklet={tracklet}
isActive={activeTracklet?.id === tracklet.id}
onClick={() => {
setActiveTrackletId(tracklet.id);
}}
/>
);
})}
{isAddObjectEnabled && <AddObjectButton />}
{isObjectLimitReached && <LimitNotice />}
</div>
<ObjectsToolbarBottomActions onTabChange={onTabChange} />
</div>
);
}

52
sam2/demo/frontend/src/common/components/annotations/ObjectsToolbarBottomActions.tsx Normal file
View File

@@ -0,0 +1,52 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ClearAllPointsInVideoButton from '@/common/components/annotations/ClearAllPointsInVideoButton';
import CloseSessionButton from '@/common/components/annotations/CloseSessionButton';
import TrackAndPlayButton from '@/common/components/button/TrackAndPlayButton';
import ToolbarBottomActionsWrapper from '@/common/components/toolbar/ToolbarBottomActionsWrapper';
import {
EFFECT_TOOLBAR_INDEX,
OBJECT_TOOLBAR_INDEX,
} from '@/common/components/toolbar/ToolbarConfig';
import {streamingStateAtom} from '@/demo/atoms';
import {useAtomValue} from 'jotai';
type Props = {
onTabChange: (newIndex: number) => void;
};
export default function ObjectsToolbarBottomActions({onTabChange}: Props) {
const streamingState = useAtomValue(streamingStateAtom);
const isTrackingEnabled =
streamingState !== 'none' && streamingState !== 'full';
function handleSwitchToEffectsTab() {
onTabChange(EFFECT_TOOLBAR_INDEX);
}
return (
<ToolbarBottomActionsWrapper>
<ClearAllPointsInVideoButton
onRestart={() => onTabChange(OBJECT_TOOLBAR_INDEX)}
/>
{isTrackingEnabled && <TrackAndPlayButton />}
{streamingState === 'full' && (
<CloseSessionButton onSessionClose={handleSwitchToEffectsTab} />
)}
</ToolbarBottomActionsWrapper>
);
}

43
sam2/demo/frontend/src/common/components/annotations/ObjectsToolbarHeader.tsx Normal file
View File

@@ -0,0 +1,43 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ToolbarHeaderWrapper from '@/common/components/toolbar/ToolbarHeaderWrapper';
import {isStreamingAtom, streamingStateAtom} from '@/demo/atoms';
import {useAtomValue} from 'jotai';
export default function ObjectsToolbarHeader() {
const isStreaming = useAtomValue(isStreamingAtom);
const streamingState = useAtomValue(streamingStateAtom);
return (
<ToolbarHeaderWrapper
title={
streamingState === 'full'
? 'Review tracked objects'
: isStreaming
? 'Tracking objects'
: 'Select objects'
}
description={
streamingState === 'full'
? 'Review your selected objects across the video, and continue to edit if needed. Once everything looks good, press “Next” to continue.'
: isStreaming
? 'Watch the video closely for any places where your objects aren’t tracked correctly. You can also stop tracking to make additional edits.'
: 'Adjust the selection of your object, or add additional objects. Press “Track objects” to track your objects throughout the video.'
}
className="mb-8"
/>
);
}

44
sam2/demo/frontend/src/common/components/annotations/PointsToggle.tsx Normal file
View File

@@ -0,0 +1,44 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {labelTypeAtom} from '@/demo/atoms';
import {AddFilled, SubtractFilled} from '@carbon/icons-react';
import {useAtom} from 'jotai';
export default function PointsToggle() {
const [labelType, setLabelType] = useAtom(labelTypeAtom);
const isPositive = labelType === 'positive';
const buttonStyle = (selected: boolean) =>
`btn-md bg-graydark-800 !text-white md:px-2 lg:px-4 py-0.5 ${selected ? `border border-white hover:bg-graydark-800` : `border-graydark-700 hover:bg-graydark-700`}`;
return (
<div className="flex items-center w-full md:ml-2">
<div className="join group grow gap-[1px]">
<button
className={`w-1/2 btn join-item text-white ${buttonStyle(isPositive)}`}
onClick={() => setLabelType('positive')}>
<AddFilled size={24} className="text-blue-500" /> Add
</button>
<button
className={`w-1/2 btn join-item text-red-700 ${buttonStyle(!isPositive)}`}
onClick={() => setLabelType('negative')}>
<SubtractFilled size={24} className="text-red-400" />
Remove
</button>
</div>
</div>
);
}

40
sam2/demo/frontend/src/common/components/annotations/PrimaryCTAButton.tsx Normal file
View File

@@ -0,0 +1,40 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import GradientBorder from '@/common/components/button/GradientBorder';
import type {ReactNode} from 'react';
type Props = {
disabled?: boolean;
endIcon?: ReactNode;
} & React.DOMAttributes<HTMLButtonElement>;
export default function PrimaryCTAButton({
children,
disabled,
endIcon,
...props
}: Props) {
return (
<GradientBorder disabled={disabled}>
<button
className={`btn ${disabled && 'btn-disabled'} !rounded-full !bg-black !text-white !border-none`}
{...props}>
{children}
{endIcon != null && endIcon}
</button>
</GradientBorder>
);
}

88
sam2/demo/frontend/src/common/components/annotations/ToolbarObject.tsx Normal file
View File

@@ -0,0 +1,88 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ObjectActions from '@/common/components/annotations/ObjectActions';
import ObjectPlaceholder from '@/common/components/annotations/ObjectPlaceholder';
import ObjectThumbnail from '@/common/components/annotations/ObjectThumbnail';
import ToolbarObjectContainer from '@/common/components/annotations/ToolbarObjectContainer';
import useVideo from '@/common/components/video/editor/useVideo';
import {BaseTracklet} from '@/common/tracker/Tracker';
import emptyFunction from '@/common/utils/emptyFunction';
import {activeTrackletObjectIdAtom} from '@/demo/atoms';
import {useSetAtom} from 'jotai';
type Props = {
label: string;
tracklet: BaseTracklet;
isActive: boolean;
isMobile?: boolean;
onClick?: () => void;
onThumbnailClick?: () => void;
};
export default function ToolbarObject({
label,
tracklet,
isActive,
isMobile = false,
onClick,
onThumbnailClick = emptyFunction,
}: Props) {
const video = useVideo();
const setActiveTrackletId = useSetAtom(activeTrackletObjectIdAtom);
async function handleCancelNewObject() {
try {
await video?.deleteTracklet(tracklet.id);
} catch (error) {
reportError(error);
} finally {
setActiveTrackletId(null);
}
}
if (!tracklet.isInitialized) {
return (
<ToolbarObjectContainer
alignItems="center"
isActive={isActive}
title="New object"
subtitle="No object is currently selected. Click an object in the video."
thumbnail={<ObjectPlaceholder showPlus={false} />}
isMobile={isMobile}
onClick={onClick}
onCancel={handleCancelNewObject}
/>
);
}
return (
<ToolbarObjectContainer
isActive={isActive}
onClick={onClick}
title={label}
subtitle=""
thumbnail={
<ObjectThumbnail
thumbnail={tracklet.thumbnail}
color={tracklet.color}
onClick={onThumbnailClick}
/>
}
isMobile={isMobile}>
<ObjectActions objectId={tracklet.id} active={isActive} />
</ToolbarObjectContainer>
);
}

123
sam2/demo/frontend/src/common/components/annotations/ToolbarObjectContainer.tsx Normal file
View File

@@ -0,0 +1,123 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {spacing} from '@/theme/tokens.stylex';
import {Close} from '@carbon/icons-react';
import stylex from '@stylexjs/stylex';
import {PropsWithChildren, ReactNode} from 'react';
const sharedStyles = stylex.create({
container: {
display: 'flex',
overflow: 'hidden',
cursor: 'pointer',
flexShrink: 0,
borderTop: 'none',
backgroundColor: {
'@media screen and (max-width: 768px)': '#000',
},
paddingHorizontal: {
default: spacing[8],
'@media screen and (max-width: 768px)': spacing[5],
},
paddingBottom: {
default: spacing[8],
'@media screen and (max-width: 768px)': 10,
},
},
activeContainer: {
background: '#000',
borderRadius: 16,
marginHorizontal: 16,
padding: {
default: spacing[4],
'@media screen and (max-width: 768px)': spacing[5],
},
marginBottom: {
default: spacing[8],
'@media screen and (max-width: 768px)': 0,
},
},
itemsCenter: {
alignItems: 'center',
},
rightColumn: {
marginStart: {
default: spacing[4],
'@media screen and (max-width: 768px)': 0,
},
flexGrow: 1,
alignItems: 'center',
},
});
type ToolbarObjectContainerProps = PropsWithChildren<{
alignItems?: 'top' | 'center';
isActive: boolean;
title: string;
subtitle: string;
thumbnail: ReactNode;
isMobile: boolean;
onCancel?: () => void;
onClick?: () => void;
}>;
export default function ToolbarObjectContainer({
alignItems = 'top',
children,
isActive,
title,
subtitle,
thumbnail,
isMobile,
onClick,
onCancel,
}: ToolbarObjectContainerProps) {
if (isMobile) {
return (
<div
onClick={onClick}
{...stylex.props(sharedStyles.container, sharedStyles.itemsCenter)}>
<div {...stylex.props(sharedStyles.rightColumn)}>{children}</div>
</div>
);
}
return (
<div
onClick={onClick}
{...stylex.props(
sharedStyles.container,
isActive && sharedStyles.activeContainer,
alignItems === 'center' && sharedStyles.itemsCenter,
)}>
{thumbnail}
<div {...stylex.props(sharedStyles.rightColumn)}>
<div className="text-md font-semibold ml-2">{title}</div>
{subtitle.length > 0 && (
<div className="text-sm text-gray-400 leading-5 mt-2 ml-2">
{subtitle}
</div>
)}
{children}
</div>
{onCancel != null && (
<div className="items-start self-stretch" onClick={onCancel}>
<Close size={32} />
</div>
)}
</div>
);
}

173
sam2/demo/frontend/src/common/components/annotations/TrackletSwimlane.tsx Normal file
View File

@@ -0,0 +1,173 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import useSelectedFrameHelper from '@/common/components/video/filmstrip/useSelectedFrameHelper';
import {BaseTracklet, DatalessMask} from '@/common/tracker/Tracker';
import {spacing, w} from '@/theme/tokens.stylex';
import stylex from '@stylexjs/stylex';
import {useMemo} from 'react';
const styles = stylex.create({
container: {
display: 'flex',
alignItems: 'center',
gap: spacing[4],
width: '100%',
},
trackletNameContainer: {
width: w[12],
textAlign: 'center',
fontSize: '10px',
color: 'white',
},
swimlaneContainer: {
flexGrow: 1,
position: 'relative',
display: 'flex',
height: 12,
marginVertical: '0.25rem' /* 4px */,
'@media screen and (max-width: 768px)': {
marginVertical: 0,
},
},
swimlane: {
position: 'absolute',
left: 0,
top: '50%',
width: '100%',
height: 1,
transform: 'translate3d(0, -50%, 0)',
opacity: 0.4,
},
segment: {
position: 'absolute',
top: '50%',
height: 1,
transform: 'translate3d(0, -50%, 0)',
},
segmentationPoint: {
position: 'absolute',
top: '50%',
transform: 'translate3d(0, -50%, 0)',
borderRadius: '50%',
cursor: 'pointer',
width: 12,
height: 12,
'@media screen and (max-width: 768px)': {
width: 8,
height: 8,
},
},
});
type SwimlineSegment = {
left: number;
width: number;
};
type Props = {
tracklet: BaseTracklet;
onSelectFrame: (tracklet: BaseTracklet, index: number) => void;
};
function getSwimlaneSegments(masks: DatalessMask[]): SwimlineSegment[] {
if (masks.length === 0) {
return [];
}
const swimlineSegments: SwimlineSegment[] = [];
let left = -1;
for (let frameIndex = 0; frameIndex < masks.length; ++frameIndex) {
const isEmpty = masks?.[frameIndex]?.isEmpty ?? true;
if (left === -1 && !isEmpty) {
left = frameIndex;
} else if (left !== -1 && (isEmpty || frameIndex == masks.length - 1)) {
swimlineSegments.push({
left,
width: frameIndex - left + 1,
});
left = -1;
}
}
return swimlineSegments;
}
export default function TrackletSwimlane({tracklet, onSelectFrame}: Props) {
const selection = useSelectedFrameHelper();
const segments = useMemo(() => {
return getSwimlaneSegments(tracklet.masks);
}, [tracklet.masks]);
const framesWithPoints = tracklet.points.reduce<number[]>(
(frames, pts, frameIndex) => {
if (pts != null && pts.length > 0) {
frames.push(frameIndex);
}
return frames;
},
[],
);
if (selection === null) {
return;
}
return (
<div {...stylex.props(styles.container)}>
<div {...stylex.props(styles.trackletNameContainer)}>
Object {tracklet.id + 1}
</div>
<div {...stylex.props(styles.swimlaneContainer)}>
<div
{...stylex.props(styles.swimlane)}
style={{
backgroundColor: tracklet.color,
}}
/>
{segments.map(segment => {
return (
<div
key={segment.left}
{...stylex.props(styles.segment)}
style={{
backgroundColor: tracklet.color,
left: selection.toPosition(segment.left),
width: selection.toPosition(segment.width),
}}
/>
);
})}
{framesWithPoints.map(index => {
return (
<div
key={`frame${index}`}
onClick={() => {
onSelectFrame?.(tracklet, index);
}}
{...stylex.props(styles.segmentationPoint)}
style={{
left: Math.floor(selection.toPosition(index) - 4),
backgroundColor: tracklet.color,
}}
/>
);
})}
</div>
</div>
);
}

55
sam2/demo/frontend/src/common/components/annotations/TrackletsAnnotation.tsx Normal file
View File

@@ -0,0 +1,55 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import TrackletSwimlane from '@/common/components/annotations/TrackletSwimlane';
import useTracklets from '@/common/components/annotations/useTracklets';
import useVideo from '@/common/components/video/editor/useVideo';
import {BaseTracklet} from '@/common/tracker/Tracker';
import {m, spacing} from '@/theme/tokens.stylex';
import stylex from '@stylexjs/stylex';
const styles = stylex.create({
container: {
marginTop: m[3],
height: 75,
paddingHorizontal: spacing[4],
'@media screen and (max-width: 768px)': {
height: 25,
},
},
});
export default function TrackletsAnnotation() {
const video = useVideo();
const tracklets = useTracklets();
function handleSelectFrame(_tracklet: BaseTracklet, index: number) {
if (video !== null) {
video.frame = index;
}
}
return (
<div {...stylex.props(styles.container)}>
{tracklets.map(tracklet => (
<TrackletSwimlane
key={tracklet.id}
tracklet={tracklet}
onSelectFrame={handleSelectFrame}
/>
))}
</div>
);
}

21
sam2/demo/frontend/src/common/components/annotations/useTracklets.ts Normal file
View File

@@ -0,0 +1,21 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {trackletObjectsAtom} from '@/demo/atoms';
import {useAtomValue} from 'jotai';
export default function useTracklets() {
return useAtomValue(trackletObjectsAtom);
}

73
sam2/demo/frontend/src/common/components/button/GradientBorder.tsx Normal file
View File

@@ -0,0 +1,73 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import stylex from '@stylexjs/stylex';
import {gradients} from '@/theme/tokens.stylex';
enum GradientTypes {
fullGradient = 'fullGradient',
bluePinkGradient = 'bluePinkGradient',
}
type Props = {
gradientType?: GradientTypes;
disabled?: boolean;
rounded?: boolean;
className?: string;
} & React.DOMAttributes<HTMLDivElement>;
const styles = stylex.create({
animationHover: {
':hover': {
backgroundPosition: '300% 100%',
},
},
fullGradient: {
border: '2px solid transparent',
background: gradients['rainbow'],
backgroundSize: '100% 400%',
transition: 'background 0.35s ease-in-out',
},
bluePinkGradient: {
border: '2px solid transparent',
background: gradients['rainbow'],
},
});
export default function GradientBorder({
gradientType = GradientTypes.fullGradient,
disabled,
rounded = true,
className = '',
children,
}: Props) {
const gradient = (name: GradientTypes) => {
if (name === GradientTypes.fullGradient) {
return styles.fullGradient;
} else if (name === GradientTypes.bluePinkGradient) {
return styles.bluePinkGradient;
}
};
return (
<div
className={`${stylex(gradient(gradientType), !disabled && styles.animationHover)} ${disabled && 'opacity-30'} ${rounded && 'rounded-full'} ${className}`}>
{children}
</div>
);
}

94
sam2/demo/frontend/src/common/components/button/PlaybackButton.tsx Normal file
View File

@@ -0,0 +1,94 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {OBJECT_TOOLBAR_INDEX} from '@/common/components/toolbar/ToolbarConfig';
import Tooltip from '@/common/components/Tooltip';
import useVideo from '@/common/components/video/editor/useVideo';
import {isPlayingAtom, streamingStateAtom, toolbarTabIndex} from '@/demo/atoms';
import {PauseFilled, PlayFilledAlt} from '@carbon/icons-react';
import {useAtomValue} from 'jotai';
import {useCallback, useEffect} from 'react';
export default function PlaybackButton() {
const tabIndex = useAtomValue(toolbarTabIndex);
const streamingState = useAtomValue(streamingStateAtom);
const isPlaying = useAtomValue(isPlayingAtom);
const video = useVideo();
const isDisabled =
tabIndex === OBJECT_TOOLBAR_INDEX &&
streamingState !== 'none' &&
streamingState !== 'full';
const handlePlay = useCallback(() => {
video?.play();
}, [video]);
const handlePause = useCallback(() => {
video?.pause();
}, [video]);
const handleClick = useCallback(() => {
if (isDisabled) {
return;
}
if (isPlaying) {
handlePause();
} else {
handlePlay();
}
}, [isDisabled, isPlaying, handlePlay, handlePause]);
useEffect(() => {
const handleKey = (event: KeyboardEvent) => {
const callback = {
KeyK: handleClick,
}[event.code];
if (callback != null) {
event.preventDefault();
callback();
}
};
document.addEventListener('keydown', handleKey);
return () => {
document.removeEventListener('keydown', handleKey);
};
}, [handleClick]);
return (
<Tooltip message={`${isPlaying ? 'Pause' : 'Play'} (k)`}>
<button
disabled={isDisabled}
className={`group !rounded-full !w-10 !h-10 flex items-center justify-center ${getButtonStyles(isDisabled)}`}
onClick={handleClick}>
{isPlaying ? (
<PauseFilled size={18} />
) : (
<PlayFilledAlt
size={18}
className={!isDisabled ? 'group-hover:text-green-500' : ''}
/>
)}
</button>
</Tooltip>
);
}
function getButtonStyles(isDisabled: boolean): string {
if (isDisabled) {
return '!bg-gray-600 !text-graydark-700';
}
return `!text-black bg-white`;
}

40
sam2/demo/frontend/src/common/components/button/PrimaryCTAButton.tsx Normal file
View File

@@ -0,0 +1,40 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import GradientBorder from '@/common/components/button/GradientBorder';
import type {ReactNode} from 'react';
type Props = {
disabled?: boolean;
endIcon?: ReactNode;
} & React.DOMAttributes<HTMLButtonElement>;
export default function PrimaryCTAButton({
children,
disabled,
endIcon,
...props
}: Props) {
return (
<GradientBorder disabled={disabled}>
<button
className={`btn ${disabled && 'btn-disabled'} !rounded-full !bg-black !text-white !border-none`}
{...props}>
{children}
{endIcon != null && endIcon}
</button>
</GradientBorder>
);
}

27
sam2/demo/frontend/src/common/components/button/ResponsiveButton.tsx Normal file
View File

@@ -0,0 +1,27 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import useScreenSize from '@/common/screen/useScreenSize';
import type {ReactNode} from 'react';
import type {ButtonProps} from 'react-daisyui';
import {Button} from 'react-daisyui';
type Props = ButtonProps & {startIcon: ReactNode};
export default function ResponsiveButton(props: Props) {
const {isMobile} = useScreenSize();
return <Button {...props}>{!isMobile && props.children}</Button>;
}

123
sam2/demo/frontend/src/common/components/button/TrackAndPlayButton.tsx Normal file
View File

@@ -0,0 +1,123 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import PrimaryCTAButton from '@/common/components/button/PrimaryCTAButton';
import useMessagesSnackbar from '@/common/components/snackbar/useDemoMessagesSnackbar';
import useFunctionThrottle from '@/common/components/useFunctionThrottle';
import useVideo from '@/common/components/video/editor/useVideo';
import {
areTrackletObjectsInitializedAtom,
isStreamingAtom,
sessionAtom,
streamingStateAtom,
} from '@/demo/atoms';
import {ChevronRight} from '@carbon/icons-react';
import {useAtom, useAtomValue, useSetAtom} from 'jotai';
import {useCallback, useEffect} from 'react';
export default function TrackAndPlayButton() {
const video = useVideo();
const [isStreaming, setIsStreaming] = useAtom(isStreamingAtom);
const streamingState = useAtomValue(streamingStateAtom);
const areObjectsInitialized = useAtomValue(areTrackletObjectsInitializedAtom);
const setSession = useSetAtom(sessionAtom);
const {enqueueMessage} = useMessagesSnackbar();
const {isThrottled, maxThrottles, throttle} = useFunctionThrottle(250, 4);
const isTrackAndPlayDisabled =
streamingState === 'aborting' || streamingState === 'requesting';
useEffect(() => {
function onStreamingStarted() {
setIsStreaming(true);
}
video?.addEventListener('streamingStarted', onStreamingStarted);
function onStreamingCompleted() {
enqueueMessage('trackAndPlayComplete');
setIsStreaming(false);
}
video?.addEventListener('streamingCompleted', onStreamingCompleted);
return () => {
video?.removeEventListener('streamingStarted', onStreamingStarted);
video?.removeEventListener('streamingCompleted', onStreamingCompleted);
};
}, [video, setIsStreaming, enqueueMessage]);
const handleTrackAndPlay = useCallback(() => {
if (isTrackAndPlayDisabled) {
return;
}
if (maxThrottles && isThrottled) {
enqueueMessage('trackAndPlayThrottlingWarning');
}
// Throttling is only applied while streaming because we should
// only throttle after a user has aborted inference. This way,
// a user can still quickly abort a stream if they notice the
// inferred mask is misaligned.
throttle(
() => {
if (!isStreaming) {
enqueueMessage('trackAndPlayClick');
video?.streamMasks();
setSession(previousSession =>
previousSession == null
? previousSession
: {...previousSession, ranPropagation: true},
);
} else {
video?.abortStreamMasks();
}
},
{enableThrottling: isStreaming},
);
}, [
isTrackAndPlayDisabled,
isThrottled,
isStreaming,
maxThrottles,
video,
setSession,
enqueueMessage,
throttle,
]);
useEffect(() => {
const handleKey = (event: KeyboardEvent) => {
const callback = {
KeyK: handleTrackAndPlay,
}[event.code];
if (callback != null) {
event.preventDefault();
callback();
}
};
document.addEventListener('keydown', handleKey);
return () => {
document.removeEventListener('keydown', handleKey);
};
}, [handleTrackAndPlay]);
return (
<PrimaryCTAButton
disabled={isThrottled || !areObjectsInitialized}
onClick={handleTrackAndPlay}
endIcon={isStreaming ? undefined : <ChevronRight size={20} />}>
{isStreaming ? 'Cancel Tracking' : 'Track objects'}
</PrimaryCTAButton>
);
}

36
sam2/demo/frontend/src/common/components/code/InitializeLocalMonaco.ts Normal file
View File

@@ -0,0 +1,36 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {loader} from '@monaco-editor/react';
import Logger from '@/common/logger/Logger';
import * as monaco from 'monaco-editor';
import editorWorker from 'monaco-editor/esm/vs/editor/editor.worker?worker';
import tsWorker from 'monaco-editor/esm/vs/language/typescript/ts.worker?worker';
self.MonacoEnvironment = {
getWorker(_, label) {
if (label === 'typescript' || label === 'javascript') {
return new tsWorker();
}
return new editorWorker();
},
};
loader.config({monaco});
loader.init().then(monaco => {
Logger.debug('initialized monaco', monaco);
});

61
sam2/demo/frontend/src/common/components/effects/BackgroundEffects.tsx Normal file
View File

@@ -0,0 +1,61 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {backgroundEffects} from '@/common/components/effects/EffectsUtils';
import EffectVariantBadge from '@/common/components/effects/EffectVariantBadge';
import ToolbarActionIcon from '@/common/components/toolbar/ToolbarActionIcon';
import ToolbarSection from '@/common/components/toolbar/ToolbarSection';
import useVideoEffect from '@/common/components/video/editor/useVideoEffect';
import {EffectIndex} from '@/common/components/video/effects/Effects';
import {activeBackgroundEffectAtom} from '@/demo/atoms';
import {useAtomValue} from 'jotai';
export default function BackgroundEffects() {
const setEffect = useVideoEffect();
const activeEffect = useAtomValue(activeBackgroundEffectAtom);
return (
<ToolbarSection title="Background" borderBottom={false}>
{backgroundEffects.map(backgroundEffect => {
return (
<ToolbarActionIcon
variant="toggle"
key={backgroundEffect.title}
icon={backgroundEffect.Icon}
title={backgroundEffect.title}
isActive={activeEffect.name === backgroundEffect.effectName}
badge={
activeEffect.name === backgroundEffect.effectName && (
<EffectVariantBadge
label={`${activeEffect.variant + 1}/${activeEffect.numVariants}`}
/>
)
}
onClick={() => {
if (activeEffect.name === backgroundEffect.effectName) {
setEffect(backgroundEffect.effectName, EffectIndex.BACKGROUND, {
variant:
(activeEffect.variant + 1) % activeEffect.numVariants,
});
} else {
setEffect(backgroundEffect.effectName, EffectIndex.BACKGROUND);
}
}}
/>
);
})}
</ToolbarSection>
);
}

41
sam2/demo/frontend/src/common/components/effects/EffectVariantBadge.tsx Normal file
View File

@@ -0,0 +1,41 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {right, top} from '@/theme/tokens.stylex';
import stylex from '@stylexjs/stylex';
const styles = stylex.create({
variantBadge: {
position: 'absolute',
top: top[1],
right: right[1],
backgroundColor: '#280578',
color: '#D2D2FF',
fontVariantNumeric: 'tabular-nums',
paddingHorizontal: 4,
paddingVertical: 1,
fontSize: 9,
borderRadius: 6,
fontWeight: 'bold',
},
});
type Props = {
label: string;
};
export default function VariantBadge({label}: Props) {
return <div {...stylex.props(styles.variantBadge)}>{label}</div>;
}

93
sam2/demo/frontend/src/common/components/effects/EffectsCarousel.tsx Normal file
View File

@@ -0,0 +1,93 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {CarouselContainerShadow} from '@/common/components/effects/EffectsCarouselShadow';
import {DemoEffect} from '@/common/components/effects/EffectsUtils';
import useVideoEffect from '@/common/components/video/editor/useVideoEffect';
import type {EffectIndex} from '@/common/components/video/effects/Effects';
import {Effects} from '@/common/components/video/effects/Effects';
import {color, fontSize, spacing} from '@/theme/tokens.stylex';
import stylex from '@stylexjs/stylex';
type Props = {
label: string;
effects: DemoEffect[];
activeEffect: keyof Effects;
index: EffectIndex;
};
const styles = stylex.create({
container: {
display: 'flex',
flexDirection: 'column',
gap: spacing[2],
width: '100%',
},
label: {
fontSize: fontSize['xs'],
color: '#A6ACB2',
textAlign: 'center',
},
carouselContainer: {
position: 'relative',
borderRadius: '8px',
overflow: 'hidden',
width: '100%',
height: '120px',
backgroundColor: color['gray-700'],
},
});
export default function EffectsCarousel({
label,
effects,
activeEffect,
index: effectIndex,
}: Props) {
const setEffect = useVideoEffect();
return (
<div {...stylex.props(styles.container)}>
<div {...stylex.props(styles.label)}>{label}</div>
<div {...stylex.props(styles.carouselContainer)}>
<CarouselContainerShadow isTop={true} />
<div className="carousel carousel-vertical w-full h-full text-white">
<div className={`carousel-item h-6`} />
{effects.map(({effectName, Icon, title}, index) => {
const isActive = activeEffect === effectName;
return (
<div
key={index}
className={`carousel-item flex items-center h-6 gap-2 px-4`}
onClick={() => setEffect(effectName, effectIndex)}>
<Icon
color={isActive ? '#FB73A5' : undefined}
size={18}
fontWeight={10}
/>
<div
className={`text-sm ${isActive ? 'text-[#FB73A5] font-bold' : 'font-medium'}`}>
{title}
</div>
</div>
);
})}
<div className={`carousel-item h-6`} />
</div>
<CarouselContainerShadow isTop={false} />
</div>
</div>
);
}

46
sam2/demo/frontend/src/common/components/effects/EffectsCarouselShadow.tsx Normal file
View File

@@ -0,0 +1,46 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {spacing} from '@/theme/tokens.stylex';
import stylex from '@stylexjs/stylex';
const styles = stylex.create({
container: {
position: 'absolute',
width: '100%',
height: spacing[8],
pointerEvents: 'none',
},
});
type CarouselContainerShadowProps = {
isTop: boolean;
};
const edgeColor = 'rgba(55, 62, 65, 1)';
const transitionColor = 'rgba(55, 62, 65, 0.2)';
export function CarouselContainerShadow({isTop}: CarouselContainerShadowProps) {
return (
<div
{...stylex.props(styles.container)}
style={{
background: `linear-gradient(${isTop ? `${edgeColor}, ${transitionColor}` : `${transitionColor}, ${edgeColor}`})`,
top: isTop ? 0 : undefined,
bottom: isTop ? undefined : 0,
}}
/>
);
}

48
sam2/demo/frontend/src/common/components/effects/EffectsToolbar.tsx Normal file
View File

@@ -0,0 +1,48 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import BackgroundEffects from '@/common/components/effects/BackgroundEffects';
import EffectsToolbarBottomActions from '@/common/components/effects/EffectsToolbarBottomActions';
import EffectsToolbarHeader from '@/common/components/effects/EffectsToolbarHeader';
import HighlightEffects from '@/common/components/effects/HighlightEffects';
import useMessagesSnackbar from '@/common/components/snackbar/useDemoMessagesSnackbar';
import {useEffect, useRef} from 'react';
type Props = {
onTabChange: (newIndex: number) => void;
};
export default function EffectsToolbar({onTabChange}: Props) {
const isEffectsMessageShown = useRef(false);
const {enqueueMessage} = useMessagesSnackbar();
useEffect(() => {
if (!isEffectsMessageShown.current) {
isEffectsMessageShown.current = true;
enqueueMessage('effectsMessage');
}
}, [enqueueMessage]);
return (
<div className="flex flex-col h-full">
<EffectsToolbarHeader />
<div className="grow overflow-y-auto">
<HighlightEffects />
<BackgroundEffects />
</div>
<EffectsToolbarBottomActions onTabChange={onTabChange} />
</div>
);
}

46
sam2/demo/frontend/src/common/components/effects/EffectsToolbarBottomActions.tsx Normal file
View File

@@ -0,0 +1,46 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import PrimaryCTAButton from '@/common/components/button/PrimaryCTAButton';
import RestartSessionButton from '@/common/components/session/RestartSessionButton';
import ToolbarBottomActionsWrapper from '@/common/components/toolbar/ToolbarBottomActionsWrapper';
import {
MORE_OPTIONS_TOOLBAR_INDEX,
OBJECT_TOOLBAR_INDEX,
} from '@/common/components/toolbar/ToolbarConfig';
import {ChevronRight} from '@carbon/icons-react';
type Props = {
onTabChange: (newIndex: number) => void;
};
export default function EffectsToolbarBottomActions({onTabChange}: Props) {
function handleSwitchToMoreOptionsTab() {
onTabChange(MORE_OPTIONS_TOOLBAR_INDEX);
}
return (
<ToolbarBottomActionsWrapper>
<RestartSessionButton
onRestartSession={() => onTabChange(OBJECT_TOOLBAR_INDEX)}
/>
<PrimaryCTAButton
onClick={handleSwitchToMoreOptionsTab}
endIcon={<ChevronRight />}>
Next
</PrimaryCTAButton>
</ToolbarBottomActionsWrapper>
);
}

62
sam2/demo/frontend/src/common/components/effects/EffectsToolbarHeader.tsx Normal file
View File

@@ -0,0 +1,62 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import ToolbarHeaderWrapper from '@/common/components/toolbar/ToolbarHeaderWrapper';
import useVideoEffect from '@/common/components/video/editor/useVideoEffect';
import {
EffectIndex,
effectPresets,
} from '@/common/components/video/effects/Effects';
import {BLUE_PINK_FILL} from '@/theme/gradientStyle';
import {MagicWandFilled} from '@carbon/icons-react';
import {useCallback, useRef} from 'react';
import {Button} from 'react-daisyui';
export default function EffectsToolbarHeader() {
const preset = useRef(0);
const setEffect = useVideoEffect();
const handleTogglePreset = useCallback(() => {
preset.current++;
const [background, highlight] =
effectPresets[preset.current % effectPresets.length];
setEffect(background.name, EffectIndex.BACKGROUND, {
variant: background.variant,
});
setEffect(highlight.name, EffectIndex.HIGHLIGHT, {
variant: highlight.variant,
});
}, [setEffect]);
return (
<ToolbarHeaderWrapper
title="Add effects"
description="Apply visual effects to your selected objects and the background. Keeping clicking the same effect for different variations."
bottomSection={
<div className="flex mt-1">
<Button
color="ghost"
size="md"
className={`font-medium bg-black !rounded-full hover:!bg-gradient-to-br ${BLUE_PINK_FILL} border-none`}
endIcon={<MagicWandFilled size={20} className="text-white " />}
onClick={handleTogglePreset}>
Surprise Me
</Button>
</div>
}
className="pb-4"
/>
);
}

76
sam2/demo/frontend/src/common/components/effects/EffectsUtils.ts Normal file
View File

@@ -0,0 +1,76 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import {Effects} from '@/common/components/video/effects/Effects';
import type {CarbonIconType} from '@carbon/icons-react';
import {
AppleDash,
Asterisk,
Barcode,
CenterCircle,
ColorPalette,
ColorSwitch,
Development,
Erase,
FaceWink,
Humidity,
Image,
Overlay,
TextFont,
} from '@carbon/icons-react';
export type DemoEffect = {
title: string;
Icon: CarbonIconType;
effectName: keyof Effects;
};
export const backgroundEffects: DemoEffect[] = [
{title: 'Original', Icon: Image, effectName: 'Original'},
{title: 'Erase', Icon: Erase, effectName: 'EraseBackground'},
{
title: 'Gradient',
Icon: ColorPalette,
effectName: 'Gradient',
},
{
title: 'Pixelate',
Icon: Development,
effectName: 'Pixelate',
},
{title: 'Desaturate', Icon: ColorSwitch, effectName: 'Desaturate'},
{title: 'Text', Icon: TextFont, effectName: 'BackgroundText'},
{title: 'Blur', Icon: Humidity, effectName: 'BackgroundBlur'},
{title: 'Outline', Icon: AppleDash, effectName: 'Sobel'},
];
export const highlightEffects: DemoEffect[] = [
{title: 'Original', Icon: Image, effectName: 'Cutout'},
{title: 'Erase', Icon: Erase, effectName: 'EraseForeground'},
{title: 'Gradient', Icon: ColorPalette, effectName: 'VibrantMask'},
{title: 'Pixelate', Icon: Development, effectName: 'PixelateMask'},
{
title: 'Overlay',
Icon: Overlay,
effectName: 'Overlay',
},
{title: 'Emoji', Icon: FaceWink, effectName: 'Replace'},
{title: 'Burst', Icon: Asterisk, effectName: 'Burst'},
{title: 'Spotlight', Icon: CenterCircle, effectName: 'Scope'},
];
export const moreEffects: DemoEffect[] = [
{title: 'Noisy', Icon: Barcode, effectName: 'NoisyMask'},
];

64
sam2/demo/frontend/src/common/components/effects/HighlightEffects.tsx Normal file
View File

@@ -0,0 +1,64 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import EffectVariantBadge from '@/common/components/effects/EffectVariantBadge';
import ToolbarActionIcon from '@/common/components/toolbar/ToolbarActionIcon';
import ToolbarSection from '@/common/components/toolbar/ToolbarSection';
import useVideoEffect from '@/common/components/video/editor/useVideoEffect';
import {EffectIndex} from '@/common/components/video/effects/Effects';
import {
activeHighlightEffectAtom,
activeHighlightEffectGroupAtom,
} from '@/demo/atoms';
import {useAtomValue} from 'jotai';
export default function HighlightEffects() {
const setEffect = useVideoEffect();
const activeEffect = useAtomValue(activeHighlightEffectAtom);
const activeEffectsGroup = useAtomValue(activeHighlightEffectGroupAtom);
return (
<ToolbarSection title="Selected Objects" borderBottom={true}>
{activeEffectsGroup.map(highlightEffect => {
return (
<ToolbarActionIcon
variant="toggle"
key={highlightEffect.title}
icon={highlightEffect.Icon}
title={highlightEffect.title}
isActive={activeEffect.name === highlightEffect.effectName}
badge={
activeEffect.name === highlightEffect.effectName && (
<EffectVariantBadge
label={`${activeEffect.variant + 1}/${activeEffect.numVariants}`}
/>
)
}
onClick={() => {
if (activeEffect.name === highlightEffect.effectName) {
setEffect(highlightEffect.effectName, EffectIndex.HIGHLIGHT, {
variant:
(activeEffect.variant + 1) % activeEffect.numVariants,
});
} else {
setEffect(highlightEffect.effectName, EffectIndex.HIGHLIGHT);
}
}}
/>
);
})}
</ToolbarSection>
);
}

115
sam2/demo/frontend/src/common/components/effects/MobileEffectsToolbar.tsx Normal file
View File

@@ -0,0 +1,115 @@
/**
* Copyright (c) Meta Platforms, Inc. and affiliates.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
import EffectsCarousel from '@/common/components/effects/EffectsCarousel';
import {backgroundEffects} from '@/common/components/effects/EffectsUtils';
import useVideoEffect from '@/common/components/video/editor/useVideoEffect';
import {
EffectIndex,
effectPresets,
} from '@/common/components/video/effects/Effects';
import {ListBoxes, MagicWand, MagicWandFilled} from '@carbon/icons-react';
import {useCallback, useRef, useState} from 'react';
import {Button} from 'react-daisyui';
import EffectsToolbarBottomActions from '@/common/components/effects/EffectsToolbarBottomActions';
import ToolbarProgressChip from '@/common/components/toolbar/ToolbarProgressChip';
import {
activeBackgroundEffectAtom,
activeHighlightEffectAtom,
activeHighlightEffectGroupAtom,
} from '@/demo/atoms';
import {BLUE_PINK_FILL} from '@/theme/gradientStyle';
import {useAtomValue} from 'jotai';
type Props = {
onTabChange: (newIndex: number) => void;
};
export default function MobileEffectsToolbar({onTabChange}: Props) {
const preset = useRef(0);
const setEffect = useVideoEffect();
const [showEffectsCarousels, setShowEffectsCarousels] = useState<boolean>();
const activeBackground = useAtomValue(activeBackgroundEffectAtom);
const activeHighlight = useAtomValue(activeHighlightEffectAtom);
const activeHighlightEffectsGroup = useAtomValue(
activeHighlightEffectGroupAtom,
);
const handleTogglePreset = useCallback(() => {
preset.current++;
const [background, highlight] =
effectPresets[preset.current % effectPresets.length];
setEffect(background.name, EffectIndex.BACKGROUND, {
variant: background.variant,
});
setEffect(highlight.name, EffectIndex.HIGHLIGHT, {
variant: highlight.variant,
});
}, [setEffect]);
return (
<div className="w-full">
{showEffectsCarousels ? (
<div className="flex gap-2 px-2 py-4 items-center p-6">
<Button
color="ghost"
className="mt-6 !px-2 !text-[#FB73A5]"
startIcon={<MagicWand size={20} />}
onClick={handleTogglePreset}
/>
<EffectsCarousel
label="Highlights"
effects={activeHighlightEffectsGroup}
activeEffect={activeHighlight.name}
index={1}
/>
<EffectsCarousel
label="Background"
effects={backgroundEffects}
activeEffect={activeBackground.name}
index={0}
/>
</div>
) : (
<div className="flex flex-col gap-6 p-6">
<div className="text-sm text-white">
<ToolbarProgressChip />
Apply visual effects to your selected objects and the background.
</div>
<div className="grid grid-cols-2 gap-2">
<Button
color="ghost"
endIcon={<MagicWandFilled size={20} />}
className={`font-bold bg-black !rounded-full !bg-gradient-to-br ${BLUE_PINK_FILL} border-none text-white`}
onClick={handleTogglePreset}>
Surprise Me
</Button>
<Button
color="ghost"
className={`font-bold bg-black !rounded-full border-none text-white`}
startIcon={<ListBoxes size={20} />}
onClick={() => setShowEffectsCarousels(true)}>
More effects
</Button>
</div>
</div>
)}
<EffectsToolbarBottomActions onTabChange={onTabChange} />
</div>
);
}

Some files were not shown because too many files have changed in this diff Show More