Video Semantic Segmentation on VSPW W2F protocol 25% warm-up ratio
51mIoUDiTTA
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| DiTTAApproach=ISS+SAM2, FPS=13.452026.04 | 51 | 66.6 | 94.5 | 92.6 | |
| TV3SApproach=VSS, FPS=5.712026.04 | 49.9 | 66.6 | 91.5 | 88 | |
| CoTTAApproach=ISS+TTA, FPS=18.482026.04 | 49.8 | 66.8 | 90.3 | 87.1 | |
| CFFM++Approach=VSS, FPS=5.852026.04 | 49.7 | 66.1 | 90.4 | 86.8 | |
| CFFMApproach=VSS, FPS=5.982026.04 | 49.5 | 66.5 | 90.8 | 87.3 | |
| Zero-shot Refine.Approach=ISS+SAM2, FPS=1.412026.04 | 49.5 | 66.4 | 94.8 | 93.1 | |
| SegFormerApproach=ISS, FPS=18.582026.04 | 49 | 66.4 | 88.3 | 84.3 | |
| TENTApproach=ISS+TTA, FPS=18.452026.04 | 48.7 | 65.6 | 91.6 | 88.9 | |
| AuxAdaptApproach=ISS+TTA, FPS=18.562026.04 | 48.7 | 65.6 | 90.2 | 86.9 |