---
license: cc-by-nc-4.0
task_categories:
- image-segmentation
tags:
- video-object-segmentation
- video-segmentation
- evaluation
- croissant
- neurips-2026
- video
pretty_name: FaVOS
size_categories:
- 10K<n<100K
---

# FaVOS

[Project page](https://aidaslab.github.io/FaVOS) | [Paper](https://arxiv.org/abs/2610.02946)

FaVOS is a video object segmentation benchmark with fractional temporal visibility. It contains 200 videos and 279 annotated objects that may disappear and reappear.

| Subset | Videos | Objects | Mean τ |
|---|---:|---:|---:|
| FaVOS-20 | 100 | 129 | 0.2051 |
| FaVOS-40 | 100 | 150 | 0.4048 |
| Full | 200 | 279 | 0.3124 |

FaVOS evaluation uses the first annotated frame as the prompt frame. Metrics are
computed on the remaining frames only.

Download the dataset:

```bash
huggingface-cli download snu-aidas/FaVOS --repo-type dataset --local-dir data
unzip data/JPEGImages.zip -d data
unzip data/Annotations.zip -d data
```

Evaluate predictions against:

```text
data/Annotations/<video_id>/<frame>.png
```

Use `favos-20.txt` and `favos-40.txt` as the evaluation splits. To compute Volumetric J&F with [eval_jnf_v.py](eval_jnf_v.py):

```bash
python data/eval_jnf_v.py --pred-root predictions --output-dir results/jf_v
```

## License

The annotations and the 67 self-collected videos are released under CC BY-NC 4.0. Frames from the 133 YFCC100M videos keep their original licenses. See [LICENSE.md](LICENSE.md).

Contact: csjihwanh@snu.ac.kr

## Citation

```bibtex
@inproceedings{hong2026favos,
  title     = {When Predicting Nothing Beats SAM 3: Revisiting Evaluation in Video Object Segmentation},
  author    = {Hong, Jihwan and Park, Woohyeon and Kim, Jaeik and Do, Jaeyoung},
  booktitle = {Advances in Neural Information Processing Systems},
  year      = {2026},
  note      = {Evaluations \& Datasets Track}
}
```