| Benchmark | Model | Rank | Results |
|---|---|---|---|
| video-instance-segmentation-on-hq-ytvis | VMT (Swin-L) | #1 | Tube-Boundary AP: 44.8 |
| video-instance-segmentation-on-hq-ytvis | VMT (R101) | #3 | Tube-Boundary AP: 32.5 |
| video-instance-segmentation-on-hq-ytvis | VMT (R50) | #4 | Tube-Boundary AP: 30.7 |