Referring Video Object Segmentation on Refer-Youtube-VOS
66.8J&F ScoreVideoGLaMM
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| VideoGLaMMTokens Retained=576 (100%)2026.05 | 66.8 | 65.4 | 68.2 | 100 | |
| LiteLVLMTokens Retained=1962026.05 | 66.5 | 65.1 | 67.9 | 99.5 | |
| SgMgYear=20232024.09 | 65.7 | 63.9 | 67.4 | — | |
| LiteLVLMTokens Retained=812026.05 | 64.2 | 60.8 | 67.6 | 96.1 | |
| VideoLISA-3.8BConfiguration=Post-optimization, Year=20242024.09 | 63.7 | 61.7 | 65.7 | — | |
| OnlineReferYear=20232024.09 | 63.5 | 61.6 | 65.5 | — | |
| ReferFormerYear=20222024.09 | 62.9 | 61.3 | 64.6 | — | |
| VisPrunerTokens Retained=1962026.05 | 62.3 | 60.6 | 63.9 | 93.2 | |
| VideoLISA-3.8BTraining=One-Token-Seg-All, Year=20242024.09 | 61.7 | 60.2 | 63.3 | — | |
| CITDYear=20212024.09 | 61.4 | 60 | 62.7 | — | |
| R2-VOSYear=20232024.09 | 61.3 | 59.6 | 63.1 | — | |
| VisPrunerTokens Retained=812026.05 | 60.1 | 58.1 | 62.2 | 89.9 | |
| TrackGPTModel Size=13B, Year=20232024.09 | 59.5 | 58.1 | 60.8 | — | |
| TrackGPTModel Size=7B, Year=20232024.09 | 56.4 | 55.3 | 57.4 | — | |
| MTTRYear=20222024.09 | 55.3 | 54 | 56.6 | — | |
| PMINet + CFBIYear=20212024.09 | 54.2 | 53 | 55.5 | — | |
| LISAModel Size=13B, Year=20232024.09 | 52.6 | 52.1 | 53 | — | |
| LISAModel Size=7B, Year=20232024.09 | 50.2 | 49.7 | 50.6 | — | |
| MLSAYear=20222024.09 | 49.7 | 48.4 | 50.9 | — | |
| LBDTYear=20222024.09 | 49.4 | 48.2 | 50.6 | — | |
| YOFOYear=20222024.09 | 48.6 | 47.5 | 49.7 | — | |
| CMPC-VYear=20212024.09 | 47.5 | 45.6 | 49.3 | — | |
| URVOSYear=20202024.09 | 47.2 | 45.2 | 49.1 | — |