Video Semantic Segmentation On Vspw
Metriken
mIoU
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | mIoU | Paper Title | Repository |
|---|---|---|---|
| DVIS++(VIT-L) | 63.8 | DVIS++: Improved Decoupled Framework for Universal Video Segmentation | |
| CFFM(MiT-B5) | 49.3 | Learning Local and Global Temporal Contexts for Video Semantic Segmentation | |
| Tube-Link(Swin-large) | 59.6 | Tube-Link: A Flexible Cross Tube Framework for Universal Video Segmentation | |
| MRCFA(MiT-B5) | 49.9 | Mining Relations among Cross-Frame Affinities for Video Semantic Segmentation | |
| UniVS(Swin-L) | 59.8 | UniVS: Unified and Universal Video Segmentation with Prompts as Queries |
0 of 5 row(s) selected.