Referring Video Segmentation on Ref-YouTube-VOS
73.9J&F ScoreMPG-SAM 2
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MPG-SAM 22025.01 | 73.9 | 71.7 | 76.1 | |
| FindTrack++N=10, fine-tuning=true2025.03 | 73.7 | 71.8 | 75.7 | |
| FindTrack++N=5, fine-tuning=true2025.03 | 73.1 | 71.2 | 75 | |
| GLEE-ProBackbone=EVA02-L2023.12 | 70.6 | 68.2 | 72.9 | |
| FindTrackN=52025.03 | 70.3 | 68.6 | 72 | |
| FindTrackN=102025.03 | 70.3 | 68.6 | 71.9 | |
| UNINEXTBackbone=ViT-H2023.03 | 70.1 | 67.6 | 72.7 | |
| UNINEXTBackbone=ViT-H2023.12 | 70.1 | 67.6 | 72.7 | |
| ReferDINOBackbone=Swin-B2025.01 | 69.3 | 67 | 71.5 | |
| ReferDINOBackbone=GroundingDINO-B2026.03 | 69.3 | 67 | 71.5 | |
| SAMWISEPublication=CVPR'252025.03 | 69.2 | 67.8 | 70.6 | |
| MUTRPublication=AAAI'242025.03 | 68.4 | 66.4 | 70.4 | |
| MUTRReference=AAAI'242025.01 | 68.4 | 66.4 | 70.4 | |
| AL-Ref-SAM2Backbone=GPT-42026.03 | 67.9 | 65.9 | 69.9 | |
| GLEE-PlusBackbone=Swin-L2023.12 | 67.7 | 65.6 | 69.7 | |
| ReferDINOBackbone=Swin-T2025.01 | 67.5 | 65.5 | 69.6 | |
| SOCPublication=NeurIPS'232025.03 | 67.3 | 65.3 | 69.3 | |
| SOCBackbone=Video-Swin-B2023.12 | 67.3 | 65.3 | 69.3 | |
| LoShBackbone=Swin-B2025.01 | 67.2 | 65.4 | 69 | |
| LoShPublication=CVPR'242025.03 | 67.2 | 65.4 | 69 | |
| LoShReference=CVPR'242025.01 | 67.2 | 65.4 | 69 | |
| DsHmpBackbone=Swin-B2025.01 | 67.1 | 65 | 69.1 | |
| DsHmpPublication=CVPR'242025.03 | 67.1 | 65 | 69.1 | |
| DsHmpReference=CVPR'242025.01 | 67.1 | 65 | 69.1 | |
| DsHmpBackbone=Video-Swin-B2026.03 | 67.1 | 65 | 69.1 | |
| TrajSegBackbone=LLaVA-7B2026.03 | 67 | 65.2 | 68.7 | |
| G-DINO-DHBackbone=Swin-B2025.01 | 66.7 | 64.5 | 69 | |
| VD-ITPublication=ECCV'242025.03 | 66.5 | 64.4 | 68.5 | |
| VD-ITReference=ECCV'242025.01 | 66.5 | 64.4 | 68.5 | |
| VD-ITBackbone=Video Diffusion2026.03 | 66.5 | 64.4 | 68.5 | |
| G-DINO-SHBackbone=Swin-B2025.01 | 66.2 | 64.2 | 68.3 | |
| UNINEXTBackbone=ConvNext-L2023.03 | 66.2 | 64 | 68.4 | |
| UNINEXTBackbone=ConvNext-L2023.12 | 66.2 | 64 | 68.4 | |
| SOCBackbone=Swin-B2025.01 | 66 | 64.1 | 67.9 | |
| SOCReference=NIPS'232025.01 | 66 | 64.1 | 67.9 | |
| TempCDReference=ICCV'232025.01 | 65.8 | 63.6 | 68 | |
| SgMgBackbone=Swin-B2025.01 | 65.7 | 63.9 | 67.4 | |
| SgMgPublication=ICCV'232025.03 | 65.7 | 63.9 | 67.4 | |
| SgMgReference=ICCV'232025.01 | 65.7 | 63.9 | 67.4 | |
| SgMgBackbone=SwinB2024.02 | 65.7 | — | — | |
| ReferFormerBackbone=Video-Swin-T2023.03 | 64.9 | 62.8 | 67 | |
| ReferFormerBackbone=Video-Swin-B2023.12 | 64.9 | 62.8 | 67 | |
| DMVSPublication=CVPR'252025.03 | 64.3 | 62.4 | 66.2 | |
| SSAPublication=CVPR'252025.03 | 64.3 | 62.2 | 66.4 | |
| SSABackbone=CLIP-B2026.03 | 64.3 | 62.2 | 66.4 | |
| G-DINO-DHBackbone=Swin-T2025.01 | 64.2 | 62.4 | 66.1 | |
| VideoChat-TPOzero-shot=true2024.12 | 63.9 | 52.3 | 75.4 | |
| G-DINO-SHBackbone=Swin-T2025.01 | 63.9 | 62 | 65.8 | |
| OnlineReferFine-tuning=true2025.01 | 63.9 | — | — | |
| VideoChat-TPOFine-tuning=true2025.01 | 63.9 | — | — | |
| VideoLISA2024.12 | 63.7 | 61.7 | 63.7 | |
| LoShBackbone=Swin-T2025.01 | 63.7 | 62 | 65.4 | |
| VideoLISAFine-tuning=true2025.01 | 63.7 | — | — | |
| VideoLISA (p)Backbone=LLaVA-Phi-3.8B2026.03 | 63.7 | 61.7 | 65.7 | |
| DsHmpBackbone=Swin-T2025.01 | 63.6 | 61.8 | 65.4 | |
| OnlineReferBackbone=Swin-L2026.03 | 63.5 | 61.6 | 65.5 | |
| HTMLBackbone=Swin-B2025.01 | 63.4 | 61.5 | 65.2 | |
| HTMLPublication=ICCV'232025.03 | 63.4 | 61.5 | 65.2 | |
| HTMLReference=ICCV'232025.01 | 63.4 | 61.5 | 65.2 | |
| VISAReference=ECCV'242025.01 | 63 | 61.4 | 64.7 | |
| OnlineReferBackbone=Swin-L2024.02 | 63 | — | — | |
| VISABackbone=Chat-UniVi-13B2026.03 | 63 | 61.4 | 64.7 | |
| ReferFormer2024.12 | 62.9 | 61.3 | 64.6 | |
| OnlineRefer2024.12 | 62.9 | 61 | 64.7 | |
| ReferFormerBackbone=Swin-B2025.01 | 62.9 | 61.3 | 64.6 | |
| UniVTGFine-tuning=true2025.01 | 62.9 | — | — | |
| ReferFormerFine-tuning=true2025.01 | 62.9 | — | — | |
| ReferFormerPublication=CVPR'222025.03 | 62.9 | 61.3 | 64.6 | |
| ReferFormerReference=CVPR'222025.01 | 62.9 | 61.3 | 64.6 | |
| OnlineReferReference=ICCV'232025.01 | 62.9 | 61 | 64.7 | |
| ReferFormerBackbone=Video-Swin-B2026.03 | 62.9 | 61.3 | 64.6 | |
| VLT+TCPublication=TPAMI'232025.03 | 62.7 | — | — | |
| SOCBackbone=Swin-T2025.01 | 62.4 | 61.1 | 63.7 | |
| SgMgBackbone=Swin-T2025.01 | 62 | 60.4 | 63.5 | |
| SgMgBackbone=SwinT2024.02 | 62 | — | — | |
| VideoLISABackbone=LLaVA-Phi-3.8B2026.03 | 61.7 | 60.2 | 63.3 | |
| VISABackbone=Chat-UniVi-7B2026.03 | 61.5 | 59.8 | 63.2 | |
| CITDBackbone=Ensemble2023.03 | 61.4 | 60 | 62.7 | |
| CITDBackbone=Ensemble2023.12 | 61.4 | 60 | 62.7 | |
| HTMLBackbone=Swin-T2025.01 | 61.2 | 59.5 | 63 | |
| UNINEXTBackbone=ResNet-502023.03 | 61.2 | 59.3 | 63 | |
| UNINEXTBackbone=ResNet-502023.12 | 61.2 | 59.3 | 63 | |
| TrackGPTBackbone=LLaVA-13B2026.03 | 59.5 | 58.1 | 60.8 | |
| ReferFormerBackbone=Swin-T2025.01 | 59.4 | 58 | 60.9 | |
| ReferFormerBackbone=ResNet-502023.03 | 58.7 | 57.4 | 60.1 | |
| ReferFormerBackbone=ResNet-502023.12 | 58.7 | 57.4 | 60.1 | |
| OnlineReferBackbone=R502024.02 | 57.3 | — | — | |
| TrackGPTBackbone=LLaVA-7B2026.03 | 56.4 | 55.3 | 57.4 | |
| MANet2022.07 | 55.63 | 54.75 | 56.51 | |
| ReferFormerBackbone=R502024.02 | 55.6 | — | — | |
| MTTRVenue=CVPR222022.07 | 55.32 | 54 | 56.64 | |
| MTTRBackbone=Video-Swin-T2023.03 | 55.3 | 54 | 56.6 | |
| MTTRBackbone=Video-Swin-T2026.03 | 55.3 | 54 | 56.6 | |
| VISABackbone=LLaVA-13B2026.03 | 54.4 | 54 | 54.8 | |
| PMINet + CFBIBackbone=Ensemble2023.03 | 54.2 | 53 | 55.5 | |
| PMINet + CFBIBackbone=Ensemble2023.12 | 54.2 | 53 | 55.5 | |
| VISABackbone=LLaVA-7B2026.03 | 53.9 | 53.4 | 54.3 | |
| LISA2024.12 | 52.6 | 52.1 | 53 | |
| LISAFine-tuning=true2025.01 | 52.6 | — | — | |
| YOFOBackbone=ResNet-502023.03 | 48.6 | 47.5 | 49.7 |