SegTrack v2 is a video segmentation dataset with full pixel-level annotations on multiple objects at each frame within each video.
Source: https://web.engr.oregonstate.edu/~lif/SegTrack2/dataset.html
Image Source: https://www.researchgate.net/publication/325842926_Semantic_Video_Segmentation_A_Review_on_Recent_Approaches
Variants: SegTrack v2, SegTrack-v2
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Video Object Segmentation | LOCATE | LOCATE: Self-supervised Object Discovery via … | 2023-08-22 |
| Unsupervised Object Segmentation | RCF (with post-processing) | Bootstrapping Objectness from Videos by … | 2023-04-17 |
| Unsupervised Object Segmentation | RCF (without post-processing) | Bootstrapping Objectness from Videos by … | 2023-04-17 |
| Unsupervised Object Segmentation | MOD | Motion-inductive Self-supervised Object Discovery in … | 2022-10-01 |
| Unsupervised Object Segmentation | TokenCut | TokenCut: Segmenting Objects in Images … | 2022-09-01 |
| Unsupervised Object Segmentation | OCLR | Segmenting Moving Objects via an … | 2022-07-05 |
| Unsupervised Object Segmentation | GWM | Guess What Moves: Unsupervised Video … | 2022-05-16 |
| Unsupervised Object Segmentation | AMD | The Emergence of Objectness: Learning … | 2021-11-11 |
Recent papers with results on this dataset: