Files
Track-Anything/README.md

27 lines
950 B
Markdown
Raw Normal View History

2023-04-12 13:21:43 +08:00
# Track-Anything
2023-04-14 23:19:28 +08:00
**Track-Anything** is an Efficient Development Toolkit for Video Object Tracking and Segmentation, based on [Segment Anything](https://github.com/facebookresearch/segment-anything) and [XMem](https://github.com/hkchengrex/XMem).
2023-04-12 13:21:43 +08:00
2023-04-14 00:10:07 +08:00
![image](https://github.com/gaomingqi/Track-Anything/blob/master/overview.png)
2023-04-12 13:21:43 +08:00
## Demo
2023-04-12 13:31:20 +08:00
2023-04-16 23:23:02 +08:00
https://user-images.githubusercontent.com/28050374/232322963-140b44a1-0b65-409a-b3fa-ce9f780aa40e.MP4
2023-04-12 13:31:20 +08:00
2023-04-14 22:21:03 +08:00
## Get Started
2023-04-16 21:02:48 +08:00
#### Linux
```bash
# Clone the repository:
git clone https://github.com/gaomingqi/Track-Anything.git
cd Track-Anything
2023-04-13 20:15:02 +00:00
2023-04-16 21:02:48 +08:00
# Install dependencies:
pip install -r requirements.txt
# Run the Caption-Anything gradio demo.
2023-04-16 21:22:38 +08:00
python app.py --device cuda:0 --sam_model_type vit_h --port 12212
2023-04-16 21:02:48 +08:00
```
2023-04-12 13:31:20 +08:00
2023-04-16 21:02:48 +08:00
## Acknowledgement
2023-04-16 21:38:31 +08:00
2023-04-12 14:03:31 +08:00
The project is based on [Segment Anything](https://github.com/facebookresearch/segment-anything) and [XMem](https://github.com/hkchengrex/XMem). Thanks for the authors for their efforts.