| LVMT: Video Mask Transformer for Long-term Video Segmentation | ✓ Link | 60.3 | | LVMT ViT-L (DINOv3) | 2026-09-29 |
| Context-Aware Video Instance Segmentation | ✓ Link | 58.5 | 56.1 | CAVIS(VIT-L) | 2024-07-03 |
| A Simple Video Segmenter by Tracking Objects Along Axial Trajectories | ✓ Link | 58.0 | | MaXTron (ConvNeXtV2-L) | 2023-11-30 |
| DVIS++: Improved Decoupled Framework for Universal Video Segmentation | ✓ Link | 58.0 | 56.0 | DVIS++(VIT-L) | 2023-12-20 |
| DVIS: Decoupled Video Instance Segmentation Framework | ✓ Link | 57.6 | 55.3 | DVIS(Swin-L) | 2023-06-06 |
| DVIS-DAQ: Improving Video Segmentation via Dynamic Anchor Queries | ✓ Link | 57.4 | | DVIS-DAQ ViT-L (online) (extra data) | 2024-03-29 |
| A Simple Video Segmenter by Tracking Objects Along Axial Trajectories | ✓ Link | 57.1 | | MaXTron (ConvNeXt-L) | 2023-11-30 |
| PMT: Plain Mask Transformer for Image and Video Segmentation with Frozen Vision Encoders | ✓ Link | 55.5 | | PMT-L (DINOv3-L frozen, online VPS) (extra data) | 2026-03-26 |
| VidEoMT: Your ViT is Secretly Also a Video Segmentation Model | ✓ Link | 55.2 | | VidEoMT | 2026-02-19 |
| Tracking Anything with Decoupled Video Segmentation | ✓ Link | 55.0 | 52.2 | DEVA (Mask2Former - SwinB) | 2023-09-07 |
| OMG-Seg: Is One Model Good Enough For All Segmentation? | ✓ Link | 53.3 | | OMG-Seg ConvNeXt-XX-Large (24 epochs) (extra data) | 2024-01-18 |
| Tube-Link: A Flexible Cross Tube Framework for Universal Video Segmentation | ✓ Link | 50.4 | 49.4 | Tube-Link(Swin-base) | 2023-03-22 |
| UniVS: Unified and Universal Video Segmentation with Prompts as Queries | ✓ Link | 49.3 | 58.2 | UniVS(Swin-L) | 2024-02-28 |
| TarViS: A Unified Approach for Target-based Video Segmentation | ✓ Link | 48.0 | 52.9 | TarViS (Swin-L) | 2023-01-06 |
| A Simple Video Segmenter by Tracking Objects Along Axial Trajectories | ✓ Link | 46.7 | | MaXTron (ResNet-50) | 2023-11-30 |
| TarViS: A Unified Approach for Target-based Video Segmentation | ✓ Link | 35.8 | 45.3 | TarViS (Swin-T) | 2023-01-06 |
| TarViS: A Unified Approach for Target-based Video Segmentation | ✓ Link | 33.5 | 43.1 | TarViS (ResNet-50) | 2023-01-06 |