Visual Object Tracking and Segmentation on VOTS 2023 (test)
0.766R (Robustness)HQTrack
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| HQTrackBackbone=InternImage-T, Mask Refiner=HQ-SAM_H2023.07 | 0.766 | 0.615 | 0.752 | 0.155 | 0.079 | 0.694 | — | — | |
| VMOSBackbone=InternImage-T2023.07 | 0.765 | 0.596 | 0.724 | 0.159 | 0.075 | 0.711 | — | — | |
| VMOSBackbone=ResNet-502023.07 | 0.759 | 0.564 | 0.693 | 0.155 | 0.086 | 0.691 | — | — | |
| VMOS + SAM_HBackbone=InternImage-T, Mask Refiner=SAM_H2023.07 | 0.757 | 0.61 | 0.751 | 0.159 | 0.084 | 0.706 | — | — | |
| READMem-STCNREADMem extension=true2023.05 | 0.66 | — | — | 0.25 | 0.09 | 0.57 | 0.42 | 56 | |
| READMem-QDMNREADMem extension=true2023.05 | 0.63 | — | — | 0.27 | 0.09 | 0.67 | 0.45 | 59 | |
| STCNsampling rate (s_r)=default2023.05 | 0.62 | — | — | 0.29 | 0.08 | 0.67 | 0.4 | 55 | |
| QDMNsampling rate (s_r)=default2023.05 | 0.62 | — | — | 0.28 | 0.1 | 0.69 | 0.44 | 59 | |
| STCNsampling rate (s_r)=502023.05 | 0.61 | — | — | 0.29 | 0.1 | 0.61 | 0.4 | 55 | |
| READMem-MiVOSREADMem extension=true2023.05 | 0.6 | — | — | 0.33 | 0.06 | 0.67 | 0.43 | 57 | |
| QDMNsampling rate (s_r)=502023.05 | 0.6 | — | — | 0.3 | 0.11 | 0.63 | 0.42 | 59 | |
| MiVOSsampling rate (s_r)=502023.05 | 0.58 | — | — | 0.35 | 0.07 | 0.67 | 0.39 | 55 | |
| MiVOSsampling rate (s_r)=default2023.05 | 0.54 | — | — | 0.41 | 0.06 | 0.75 | 0.38 | 55 |