Referring Video Object Segmentation on MeViS
51.3J&F ScoreGLUSA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GLUSATotal Params=7B2026.06 | 51.3 | 48.5 | 54.2 | |
| GLUSSTotal Params=7B2026.06 | 50.3 | 47.5 | 53.2 | |
| GeoLaVTotal Params=202M2026.06 | 50 | 47.4 | 52.9 | |
| SAMWISETotal Params=202M2026.06 | 49.5 | 46.6 | 52.4 | |
| ReferDINOTotal Params=230M2026.06 | 49.3 | 44.7 | 53.9 | |
| SSATotal Params=474M2026.06 | 48.6 | 44 | 53.2 | |
| SAMWISETotal Params=150M2026.06 | 48.3 | 45.4 | 51.2 | |
| EVIS-1BReference=-, Backbone=InternVL2-1B2026.06 | 46.8 | 43.7 | 49.9 | |
| EVIS†Reference=-, Backbone=Swin-T2026.06 | 46.7 | 43.5 | 49.9 | |
| DsHmpTotal Params=272M2026.06 | 46.4 | 43 | 49.8 | |
| DsHmpReference=CVPR’24, Backbone=Swin-T/Video-Swin-T2026.06 | 46.4 | 43 | 49.8 | |
| VideoGLaMMTotal Params=3.8B2026.06 | 45.2 | 42.1 | 48.2 | |
| VISA-13BTotal Params=13B2026.06 | 44.5 | 41.8 | 47.1 | |
| VideoLISA-3.8BReference=NeurIPS’24, Backbone=LLaVA-Phi-3-3.8B2026.06 | 44.4 | 41.3 | 47.6 | |
| VISA-7BTotal Params=7B2026.06 | 43.5 | 40.7 | 46.3 | |
| VISA-7BReference=ECCV’24, Backbone=Chat-UniVi-7B2026.06 | 43.5 | 40.7 | 46.3 | |
| One-Token-Seg-AllTotal Params=3.8B2026.06 | 42.3 | 39.4 | 45.2 | |
| TrackGPT-7BReference=arXiv’23, Backbone=LLaVA-7B2026.06 | 40.1 | 37.6 | 42.6 | |
| GroundingDINOTotal Params=240M, Evaluation integration=SAM22026.06 | 37.7 | 34.9 | 40.5 | |
| LISATotal Params=7B2026.06 | 37.2 | 35.1 | 39.4 | |
| LMPMTotal Params=195M2026.06 | 37.2 | 34.2 | 40.2 | |
| LMPMReference=ICCV’23, Backbone=Swin-T2026.06 | 37.2 | 34.2 | 40.2 | |
| LISA-7BReference=CVPR’23, Backbone=LLaVA-7B2026.06 | 37.2 | 35.1 | 39.4 | |
| VLT+TCReference=TPAMI’23, Backbone=Video-Swin-B2026.06 | 35.5 | 33.6 | 37.3 | |
| OnlineReferTotal Params=232M2026.06 | 32.3 | 31.5 | 33.1 | |
| ReferFormerTotal Params=237M2026.06 | 31 | 29.8 | 32.2 | |
| ReferFormerReference=CVPR’22, Backbone=Video-Swin-B2026.06 | 31 | 29.8 | 32.2 | |
| MTTRTotal Params=-2026.06 | 30 | 28.8 | 31.2 | |
| MTTRReference=CVPR’22, Backbone=Video-Swin-T2026.06 | 30 | 28.8 | 31.2 | |
| LBDTReference=CVPR’22, Backbone=ResNet-502026.06 | 29.3 | 27.8 | 30.8 | |
| URVOSReference=ECCV’20, Backbone=ResNet-502026.06 | 27.8 | 25.7 | 29.9 |