Video Object Segmentation on YouTube-VOS (val)
79.7J Score (Seen)STM
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| STM2019.04 | 79.7 | 72.8 | 84.2 | 80.9 | 79.4 | — | — | — | — | |
| Official STMExtra data=true, OL (Online learning)=false, FPS=6.32020.10 | 79.7 | 84.2 | 72.8 | 80.9 | — | 79.4 | — | — | — | |
| STMFT=false, PP=false2020.12 | 79.7 | 72.8 | 84.2 | 80.9 | 79.4 | — | — | — | — | |
| PTSNetOnline Fine-tuning=true2019.07 | 73.5 | 64.3 | — | — | — | 71.6 | — | — | — | |
| STG-Netonline learning=false, FPS=62020.12 | 72.7 | 69.1 | 75.2 | 74.9 | — | 73 | — | — | — | |
| STM-cycle+GCExtra data=true, OL (Online learning)=true, FPS=13.82020.10 | 72.2 | 62.8 | 76.3 | 71.9 | — | 70.8 | — | — | — | |
| STM-cycleExtra data=false, OL (Online learning)=false, FPS=432020.10 | 71.7 | 61.4 | 75.8 | 70.4 | — | 69.9 | — | — | — | |
| e-OSVOS-50-OnAFT=true, PP=false2020.12 | 71.7 | 74.3 | 66 | 73.8 | 71.4 | — | — | — | — | |
| BoLTVOS2019.04 | 71.6 | 64.3 | — | — | 71.1 | — | — | — | — | |
| PreMVOS2019.04 | 71.4 | 56.5 | 75.9 | 63.7 | 66.9 | — | — | — | — | |
| PREMVOSFT=true, PP=false, ensemble=true2020.12 | 71.4 | 56.5 | 75.9 | 63.7 | 66.9 | — | — | — | — | |
| AGSS-VOSExtra data=true, OL (Online learning)=false, FPS=12.52020.10 | 71.3 | 65.5 | 75.2 | 73.1 | — | 71.3 | — | — | — | |
| AGSSonline learning=false, FPS=122020.12 | 71.3 | 65.5 | 75.2 | 73.1 | — | 71.3 | — | — | — | |
| S2SOnline Fine-tuning=true2019.07 | 71 | 55.5 | — | — | — | 64.4 | — | — | — | |
| S2S2019.04 | 71 | 55.5 | 70 | 61.2 | 64.4 | — | — | — | — | |
| S2SOnline Learning=true, Backbone=ResNet-502019.09 | 71 | 55.5 | 70 | 61.2 | — | 64.4 | — | — | — | |
| S2SExtra data=false, OL (Online learning)=true, FPS=0.062020.10 | 71 | 55.5 | 70 | 61.2 | — | 64.4 | — | — | — | |
| S2SFT=false, PP=false2020.12 | 71 | 55.5 | 70 | 61.2 | 64.4 | — | — | — | — | |
| S2Sonline learning=true, FPS=0.12020.12 | 71 | 55.5 | 70 | 61.2 | — | 64.4 | — | — | — | |
| PTSNetOnline Fine-tuning=false2019.07 | 69.1 | 53.5 | — | — | — | 63.2 | — | — | — | |
| A-GAME2019.04 | 67.8 | 60.8 | — | — | 66.1 | — | — | — | — | |
| A-GAMEFT=false, PP=false2020.12 | 67.8 | 60.8 | — | — | 66.1 | — | — | — | — | |
| CapsuleVOSOnline Learning=No, Speed (frames/s)=13.52019.09 | 67.3 | 53.7 | 68.1 | 59.9 | 62.3 | — | — | — | — | |
| AGAMEonline learning=false2020.12 | 66.9 | 61.2 | — | — | — | 66 | — | — | — | |
| S2SOnline Fine-tuning=false2019.07 | 66.7 | 48.2 | — | — | — | 57.6 | — | — | — | |
| S2S (offline)Online Learning=No, Speed (frames/s)=6.252019.09 | 66.7 | 48.2 | 65.5 | 50.3 | 57.6 | — | — | — | — | |
| S2SOnline Learning=false, Backbone=ResNet-502019.09 | 66.7 | 48.2 | 65.5 | 50.3 | — | 57.7 | 6 | — | — | |
| S2SExtra data=false, OL (Online learning)=false, FPS=62020.10 | 66.7 | 48.2 | 65.5 | 50.3 | — | 57.6 | — | — | — | |
| S2Sonline learning=false, FPS=62020.12 | 66.7 | 48.2 | 65.5 | 50.3 | — | 57.7 | — | — | — | |
| RVOS2019.04 | 63.6 | 45.5 | 67.2 | 51 | 56.8 | — | — | — | — | |
| RVOSOnline Learning=false, Backbone=ResNet-502019.09 | 63.6 | 45.5 | 67.2 | 51 | — | 56.8 | 24 | — | — | |
| RVOSExtra data=false, OL (Online learning)=false, FPS=242020.10 | 63.6 | 45.5 | 67.2 | 51 | — | 56.8 | — | — | — | |
| RVOSFT=false, PP=false2020.12 | 63.6 | 45.5 | 67.2 | 51 | 56.8 | — | — | — | — | |
| RVOSonline learning=false, FPS=242020.12 | 63.6 | 45.5 | 67.2 | 51 | — | 56.8 | — | — | — | |
| DMM-Net+Online Learning=true, Backbone=ResNet-1012019.09 | 60.3 | 50.6 | 63.5 | 57.4 | — | 58 | — | — | — | |
| DMM-NetExtra data=true, OL (Online learning)=true2020.10 | 60.3 | 50.6 | 63.5 | 57.4 | — | 58 | — | — | — | |
| DMMonline learning=true2020.12 | 60.3 | 50.6 | 63.5 | 57.4 | — | 58 | — | — | — | |
| SiamMaskFine-tuning=false, Mask initialization=false, Speed (fps)=552018.12 | 60.2 | 45.1 | 58.2 | 47.7 | 52.8 | — | — | — | — | |
| SiamMaskOnline Learning=false, Backbone=ResNet-502019.09 | 60.2 | 45.1 | 58.2 | 47.7 | — | 52.8 | 55 | — | — | |
| SiamMaskonline learning=false, FPS=552020.12 | 60.2 | 45.1 | 58.2 | 47.7 | — | 52.8 | — | — | — | |
| OnAVOSFine-tuning=true, Mask initialization=true, Speed (fps)=0.12018.12 | 60.1 | 46.6 | 62.7 | 51.4 | 55.2 | — | — | — | — | |
| OnAVOSOnline Fine-tuning=true2019.07 | 60.1 | 46.6 | — | — | — | 55.2 | — | — | — | |
| OnAVOS2019.04 | 60.1 | 46.6 | 62.7 | 51.4 | 55.2 | — | — | — | — | |
| OnAVOSOnline Learning=true, Backbone=ResNet-502019.09 | 60.1 | 46.6 | 62.7 | 51.4 | — | 55.2 | — | — | — | |
| OnAVOSExtra data=true, OL (Online learning)=true, FPS=0.052020.10 | 60.1 | 46.6 | 62.7 | 51.4 | — | 55.2 | — | — | — | |
| OnAVOSFT=true, PP=true2020.12 | 60.1 | 46.6 | 62.7 | 51.4 | 55.2 | — | — | — | — | |
| OnAVOSonline learning=true, FPS=0.12020.12 | 60.1 | 46.6 | 62.7 | 51.4 | — | 55.2 | — | — | — | |
| OSMNFine-tuning=false, Mask initialization=true, Speed (fps)=8.02018.12 | 60 | 40.6 | 60.1 | 44 | 51.2 | — | — | — | — | |
| OSMNOnline Fine-tuning=false2019.07 | 60 | 40.6 | — | — | — | 51.2 | — | — | — | |
| OSMN2019.04 | 60 | 40.6 | 60.1 | 44 | 51.2 | — | — | — | — | |
| OSMNOnline Learning=No, Speed (frames/s)=7.142019.09 | 60 | 40.6 | 60.1 | 44 | 51.2 | — | — | — | — | |
| OSMNOnline Learning=false, Backbone=ResNet-502019.09 | 60 | 40.6 | 60.1 | 44 | — | 51.2 | 8 | — | — | |
| OSMNFT=false, PP=false2020.12 | 60 | 40.6 | 60.1 | 44 | 51.2 | — | — | — | — | |
| OSMNonline learning=false, FPS=82020.12 | 60 | 40.6 | 60.1 | 44 | — | 51.2 | — | — | — | |
| MaskTrackOnline Fine-tuning=false2019.07 | 59.9 | 45 | — | — | — | 53.1 | — | — | — | |
| MSK2019.04 | 59.9 | 45 | 59.5 | 47.9 | 53.1 | — | — | — | — | |
| MaskTrackExtra data=true, OL (Online learning)=true, FPS=0.052020.10 | 59.9 | 45 | 59.5 | 47.9 | — | 53.1 | — | — | — | |
| MSKFT=false, PP=false2020.12 | 59.9 | 45 | 59.5 | 47.9 | 53.1 | — | — | — | — | |
| OSVOSFine-tuning=true, Mask initialization=true, Speed (fps)=0.12018.12 | 59.8 | 54.2 | 60.5 | 60.7 | 58.8 | — | — | — | — | |
| OSVOSOnline Fine-tuning=true2019.07 | 59.8 | 54.2 | — | — | — | 58.8 | — | — | — | |
| OSVOS2019.04 | 59.8 | 54.2 | 60.5 | 60.7 | 58.8 | — | — | — | — | |
| OSVOSOnline Learning=Yes, Speed (frames/s)=0.102019.09 | 59.8 | 54.2 | 60.5 | 60.7 | 58.8 | — | — | — | — | |
| OSVOSOnline Learning=true, Backbone=ResNet-502019.09 | 59.8 | 54.2 | 60.5 | 60.7 | — | 58.8 | — | — | — | |
| OSVOSExtra data=true, OL (Online learning)=true2020.10 | 59.8 | 54.2 | 60.5 | 60.7 | — | 58.8 | — | — | — | |
| OSVOSFT=true, PP=true2020.12 | 59.8 | 54.2 | 60.5 | 60.7 | 58.8 | — | — | — | — | |
| OSVOSonline learning=true, FPS=0.12020.12 | 59.8 | 54.2 | 60.5 | 60.7 | — | 58.8 | — | — | — | |
| RGMPOnline Fine-tuning=false2019.07 | 59.5 | 45.2 | — | — | — | 53.8 | — | — | — | |
| RGMP2019.04 | 59.5 | 45.2 | — | — | 53.8 | — | — | — | — | |
| RGMPOnline Learning=false, Backbone=ResNet-502019.09 | 59.5 | — | 45.2 | — | — | 53.8 | 7 | — | — | |
| RGMPExtra data=true, OL (Online learning)=false, FPS=72020.10 | 59.5 | — | 45.2 | — | — | 53.8 | — | — | — | |
| RGMPFT=false, PP=false2020.12 | 59.5 | 45.2 | — | — | 53.8 | — | — | — | — | |
| RGMPonline learning=false, FPS=72020.12 | 59.5 | — | — | — | — | 53.8 | — | — | — | |
| DMM-NetOnline Learning=false, Backbone=ResNet-502019.09 | 59.2 | 47.6 | 62.6 | 53.9 | — | 55.8 | — | — | — | |
| DMM-Net+Online Learning=false, Backbone=ResNet-1012019.09 | 58.3 | 41.6 | 60.7 | 46.3 | — | 51.7 | 12 | — | — | |
| DMM-NetExtra data=true, OL (Online learning)=false, FPS=122020.10 | 58.3 | 41.6 | 60.7 | 46.3 | — | 51.7 | — | — | — | |
| DMMonline learning=false, FPS=122020.12 | 58.3 | 41.6 | 60.7 | 46.3 | — | 51.7 | — | — | — | |
| XuAnnotation Percentage=100%, Supervised=true2024.12 | 55.7 | 39.6 | 55.2 | 41.3 | 47.9 | — | — | — | — | |
| Stable Mean TeacherAnnotation Percentage=10%2024.12 | 48.2 | 35 | 46.7 | 35.4 | 41.3 | — | — | — | — | |
| Stable Mean TeacherAnnotation Percentage=5%2024.12 | 45.3 | 32 | 43.2 | 32.2 | 38.2 | — | — | — | — | |
| Kumar et al.Annotation Percentage=10%2024.12 | 43.1 | 31.4 | 40.8 | 31.8 | 36.8 | — | — | — | — | |
| XuAnnotation Percentage=10%, Supervised=true2024.12 | 11.6 | 10.1 | 9.6 | 9.2 | 10.1 | — | — | — | — | |
| ColorizationEvaluation protocol=one-shot/semi-supervised, Training supervision=self-supervised, Inference mask provision=first-frame mask only2026.07 | — | — | — | — | — | 37 | — | 37.4 | 36.6 | |
| CorrFlow2026.07 | — | — | — | — | 48.9 | 50.2 | — | 51.5 | — | |
| CorrFlowEvaluation protocol=one-shot/semi-supervised, Training supervision=self-supervised, Inference mask provision=first-frame mask only2026.07 | — | — | — | — | — | 44.7 | — | 45.6 | 43.8 | |
| CTC²2026.07 | — | — | — | — | 56.3 | 57 | — | 57.7 | — | |
| CTC²Evaluation protocol=one-shot/semi-supervised, Training supervision=self-supervised, Inference mask provision=first-frame mask only2026.07 | — | — | — | — | — | 62.4 | — | 65.7 | 59.1 | |
| MASTEvaluation protocol=one-shot/semi-supervised, Training supervision=self-supervised, Inference mask provision=first-frame mask only2026.07 | — | — | — | — | — | 64 | — | 67.7 | 60.3 | |
| MSK2026.07 | — | — | — | — | 51.8 | 52.8 | — | 53.8 | — | |
| OSMN2026.07 | — | — | — | — | 48.5 | 51.2 | — | 53.9 | — | |
| OSVOSEvaluation protocol=one-shot/semi-supervised, Training supervision=self-supervised, Inference mask provision=first-frame mask only2026.07 | — | — | — | — | — | 57.4 | — | 60.7 | 54.2 | |
| RGMP2026.07 | — | — | — | — | 53.9 | 53.8 | — | 53.7 | — | |
| Video Colorization2026.07 | — | — | — | — | 31.4 | 32.9 | — | 34.5 | — |