Video Object Segmentation on VOST (val)
94.2J ScoreDeAOT
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| DeAOTFPS=16.03, VRAM (GB)=10.422025.12 | 94.2 | — | — | 95.9 | 95.1 | |
| AOTFPS=15.71, VRAM (GB)=9.312025.12 | 82.2 | — | — | 98.3 | 90.2 | |
| Fast SAM2FPS=22.3, VRAM (GB)=1.18, Refinements=32025.12 | 71.9 | — | — | 98.1 | 85 | |
| CutieFPS=18.7, VRAM (GB)=8.342025.12 | 68.6 | — | — | 97.7 | 83.1 | |
| XMemFPS=18.5, VRAM (GB)=1.522025.12 | 67.4 | — | — | 97 | 82.2 | |
| SAM2FPS=14.6, VRAM (GB)=1.84, Refinements=42025.12 | 58.2 | — | — | 96.7 | 77.5 | |
| DeAOTBackbone=SwinB-DeAOT-L2024.12 | 45.6 | 33.2 | 35.1 | — | — | |
| SAM22024.12 | 44.1 | 28 | 32.7 | — | — | |
| ReVOS2024.12 | 41 | 25.3 | 31.7 | — | — | |
| Cutie-base2024.12 | 40.8 | 25.1 | 31.6 | — | — | |
| RMem + DeAOTBackbone=R50-DeAOT-L, Memory bank=restricted, Temporal position embedding=false2024.12 | 40.5 | 25.3 | 31.9 | — | — | |
| XMem2024.12 | 36 | 24.8 | 26.6 | — | — | |
| STMFPS=9.25, VRAM (GB)=7.412025.12 | 32.7 | — | — | 97.6 | 65.1 |