| video-instance-segmentation-on-hq-ytvis | SeqFormer (Swin-L) | #2 | Tube-Boundary AP: 43.3 |
| video-instance-segmentation-on-youtube-vis-validation | SeqFormer (Swin-L) | #13 | mask AP: 59.3AP50: 82.1AP75: 66.4AR1: 51.7AR10: 64.4 |
| video-instance-segmentation-on-youtube-vis-validation | SeqFormer (ResNet-101) | #19 | mask AP: 49.0AP50: 71.1AP75: 55.7AR1: 46.8AR10: 56.9 |
| video-instance-segmentation-on-youtube-vis-validation | SeqFormer (ResNet-50) | #21 | mask AP: 47.4AP50: 69.8AP75: 51.8AR1: 45.5AR10: 54.8 |
| video-instance-segmentation-on-youtube-vis-validation | SeqFormer (ResNet-50) | #24 | mask AP: 45.1AP50: 66.9AP75: 50.5AR1: 45.6AR10: 54.6 |