| video-instance-segmentation-on-ovis-1 | NOVIS (ResNet-50) | – | mask AP: 32.7AP50: 56.2AP75: 32.6AR1: 15.7AR10: 37.1 |
| video-instance-segmentation-on-ovis-1 | NOVIS (Swin-L) | – | mask AP: 43.5AP50: 68.3AP75: 43.8AR1: 19.4AR10: 46.9 |
| video-instance-segmentation-on-youtube-vis-1 | NOVIS (ResNet-50) | – | mask AP: 52.8AP50: 75.7AP75: 56.9AR1: 50.3AR10: 60.6 |
| video-instance-segmentation-on-youtube-vis-2 | NOVIS (ResNet-50) | – | mask AP: 47.2AP50: 69.4AP75: 50.0AR1: 41.3AR10: 54.4 |
| video-instance-segmentation-on-youtube-vis-2 | NOVIS (Swin-L) | – | mask AP: 59.8AP50: 82.0AP75: 66.5AR1: 47.9AR10: 64.4 |