Video Instance Segmentation on OVIS (test val)
19.6APVITA
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VITASupervision=Fully-supervised, Mask ann.=I+V, Backbone=ResNet-502023.03 | 19.6 | 41.2 | 17.4 | 11.7 | 26 | |
| VMTSupervision=Fully-supervised, Mask ann.=I+V, Backbone=ResNet-502023.03 | 16.9 | 36.4 | 13.7 | 10.4 | 22.7 | |
| VITA + MaskFreeVISSupervision=Video Mask-free, Mask ann.=I, Backbone=ResNet-502023.03 | 15.7 | 35.1 | 13.1 | 10.1 | 20.4 | |
| VITA + MaskFreeVISSupervision=Mask-free, Mask ann.=None, Backbone=ResNet-502023.03 | 13.5 | 32.7 | 10.6 | 8.8 | 18.5 | |
| VITA + BoxInstSupervision=Video Mask-free, Mask ann.=I, Backbone=ResNet-502023.03 | 12.1 | 28.3 | 10.2 | 8.8 | 17.9 | |
| VITA + BoxInstSupervision=Mask-free, Mask ann.=None, Backbone=ResNet-502023.03 | 10.3 | 27.2 | 8.4 | 7.3 | 16.2 |