Video Object Segmentation on MOSE
79J&F ScoreSENTRY-HiM-L
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SENTRY-HiM-LEvaluation Protocol=first-frame-mask2026.06 | 79 | — | — | |
| SENTRY-SA-LEvaluation Protocol=first-frame-mask2026.06 | 78.6 | — | — | |
| SENTRY-D4S-LEvaluation Protocol=first-frame-mask2026.06 | 78.4 | — | — | |
| SENTRY-SR-LEvaluation Protocol=first-frame-mask2026.06 | 77.2 | — | — | |
| HiM2SAM-LEvaluation Protocol=first-frame-mask2026.06 | 77 | — | — | |
| SAMITE-LEvaluation Protocol=first-frame-mask2026.06 | 76.8 | — | — | |
| DAM4SAM-LEvaluation Protocol=first-frame-mask2026.06 | 76.7 | — | — | |
| SAMURAI-LEvaluation Protocol=first-frame-mask2026.06 | 76 | — | — | |
| SENTRY-S2-LEvaluation Protocol=first-frame-mask2026.06 | 75.8 | — | — | |
| SAM2.1LongModel Scale=L2024.10 | 75.2 | — | — | |
| SAM2Long-LEvaluation Protocol=first-frame-mask2026.06 | 75.2 | — | — | |
| SAM 2.1Model Scale=L2024.10 | 74.5 | — | — | |
| SAM2-LEvaluation Protocol=first-frame-mask2026.06 | 74.5 | — | — | |
| Cutie-basewith STM=true, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS + MOSE2024.11 | 68.3 | 64.2 | 72.3 | |
| Cutie-smallwith STM=true, Backbone=ResNet-18, Training Data=YouTube VOS + DAVIS + MOSE2024.11 | 67.4 | 63.1 | 71.7 | |
| DEVAtraining_data=with OVIS, FPS=25.32023.09 | 66.5 | 62.3 | 70.8 | |
| LiVOSwith STM=false, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS + MOSE2024.11 | 64.8 | 60.9 | 68.7 | |
| Cutie-basewith STM=true, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 64 | 60 | 67.9 | |
| Cutie-smallwith STM=true, Backbone=ResNet-18, Training Data=YouTube VOS + DAVIS2024.11 | 62.2 | 58.2 | 66.2 | |
| DEVAtraining_data=without OVIS, FPS=25.32023.09 | 60 | 55.8 | 64.3 | |
| DEVAwith STM=true, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 60 | 55.8 | 64.3 | |
| LiVOSwith STM=false, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 59.2 | 55.6 | 62.8 | |
| DeAOTwith STM=true, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 59 | 54.6 | 63.4 | |
| AOT-R50Backbone=ResNet-50, FPS=6.42023.09 | 58.4 | 54.3 | 62.6 | |
| AOTwith STM=true, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 58.4 | 54.3 | 62.6 | |
| LiVOSwith STM=false, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS, Training iterations=30k additional2024.11 | 58.4 | 54.7 | 62 | |
| XMemvenue=ECCV'22, Training Data=with video data2023.04 | 57.6 | 53.3 | 62 | |
| XMemFPS=22.62023.09 | 56.3 | 52.1 | 60.6 | |
| XMemwith STM=true, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 56.3 | 52.1 | 60.6 | |
| Cutie-basewith STM=false, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS + MOSE, Memory frames=12024.11 | 52.6 | 48.6 | 56.6 | |
| STCNFPS=13.22023.09 | 52.5 | 48.5 | 56.6 | |
| STCNwith STM=true, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 52.5 | 48.5 | 56.6 | |
| Cutie-smallwith STM=false, Backbone=ResNet-18, Training Data=YouTube VOS + DAVIS + MOSE, Memory frames=12024.11 | 51.7 | 47.7 | 55.7 | |
| SWEMvenue=CVPR'22, Training Data=with video data2023.04 | 50.9 | 46.8 | 54.9 | |
| Cutie-basewith STM=false, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS, Memory frames=12024.11 | 50.6 | 46.6 | 54.6 | |
| Cutie-smallwith STM=false, Backbone=ResNet-18, Training Data=YouTube VOS + DAVIS, Memory frames=12024.11 | 49.3 | 45.4 | 53.2 | |
| RDEvenue=CVPR'22, Training Data=with video data2023.04 | 48.8 | 44.6 | 52.9 | |
| RDEwith STM=false, Backbone=ResNet-50, Training Data=YouTube VOS + DAVIS2024.11 | 46.8 | 42.4 | 51.3 | |
| SegGPTvenue=this work, Training Data=without video data2023.04 | 45.1 | 42.2 | 48 | |
| Paintervenue=CVPR'23, Training Data=without video data2023.04 | 14.5 | 10.4 | 18.5 |