Video Instance Segmentation on Youtube-VIS 2022 Validation
Metric: mAP_L (higher is better)
Metric: mAP_L (higher is better)
| # | Model↕ | mAP_L▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | DVIS++(VIT-L) | 50.9 | Yes | DVIS++: Improved Decoupled Framework for Univers... | 2023-12-20 | Code |
| 2 | CAVIS (VIT-L) | 48.6 | Yes | Context-Aware Video Instance Segmentation | 2024-07-03 | Code |
| 3 | CTVIS (Swin-L) | 46.4 | Yes | CTVIS: Consistent Training for Online Video Inst... | 2023-07-24 | Code |
| 4 | DVIS(Swin-L) | 45.9 | Yes | DVIS: Decoupled Video Instance Segmentation Fram... | 2023-06-06 | Code |
| 5 | CTVIS (ResNet-50) | 39.4 | Yes | CTVIS: Consistent Training for Online Video Inst... | 2023-07-24 | Code |
| 6 | InstanceFormer (Swin) | 26.3 | Yes | InstanceFormer: An Online Video Instance Segment... | 2022-08-22 | Code |
| 7 | InstanceFormer (Resnet-50) | 24.8 | Yes | InstanceFormer: An Online Video Instance Segment... | 2022-08-22 | Code |