# For CUDA
uv sync --extra cu124
# For CPU
uv sync --extra cpuuv run checkpoints/download_models.pyRun the Demo (Access at http://127.0.0.1:8080 )
gradio demo.pyuv run annotation.py# Clone the repository:
git clone https://github.com/lnikioffic/track-anything-annotate.git
cd track-anything-annotate
# Install dependencies:
pip install -r requirements.txt --index-url https://download.pytorch.org/whl/cu124
# Download Models
python checkpoints/download_models.py
# Run the gradio demo.
python demo.py
# Dataset Creation
python annotation.py# Clone the repository:
git clone https://github.com/lnikioffic/track-anything-annotate.git
cd track-anything-annotate
# Install dependencies:
pip install -r requirements.txt
# Download Models
python checkpoints/download_models.py
# Run the gradio demo.
python demo.py
# Dataset Creation
python annotation.pyIf you use this project in your work, please cite the paper:
@article{ivanov2025track,
title={Track Anything Annotate: Video annotation and dataset generation of computer vision models},
author={Ivanov, Nikita and Klimov, Mark and Glukhikh, Dmitry and Chernysheva, Tatiana and Glukhikh, Igor},
journal={arXiv preprint arXiv:2505.17884},
year={2025}
}