skareer's picture
Update README.md
5f1f202 verified
Optical flows associated with our work "We're Not Using Videos Effectively: An Updated Domain Adaptive Video Segmentation Baseline"
See the [github](https://github.com/SimarKareer/UnifiedVideoDA) for full instructions, but to install run
```bash
git lfs install
git clone https://huggingface.co/datasets/hoffman-lab/Unified-VideoDA-Generated-Flows
```