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

[New Feature] Support SAM 2.1 (#59)

Browse Source 
* support sam 2.1

* refine config path and ckpt path

* update README
This commit is contained in:
Ren Tianhe
2024-10-10 14:55:50 +08:00
committed by GitHub
parent e899ad99e8
commit 82e503604f
340 changed files with 39100 additions and 608 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
README.md
View File

@@ -14,6 +14,7 @@ Grounded SAM 2 does not introduce significant methodological changes compared to
## News
- `2024/10/10`: Support `SAM-2.1` models, if you want to use `SAM 2.1` model, you need to update to the latest code and reinstall SAM 2 follow [SAM 2.1 Installation](https://github.com/facebookresearch/sam2?tab=readme-ov-file#latest-updates).
- `2024/08/31`: Support `dump json results` in Grounded SAM 2 Image Demos (with Grounding DINO).
- `2024/08/20`: Support **Florence-2 SAM 2 Image Demo** which includes `dense region caption`, `object detection`, `phrase grounding`, and cascaded auto-label pipeline `caption + phrase grounding`.
- `2024/08/09`: Support **Ground and Track New Object** throughout the whole videos. This feature is still under development now. Credits to [Shuo Shen](https://github.com/ShuoShenDe).