One-shot Video Object Segmentation on DAVIS 2016 (val)
91.1J MeanSwinB-DeAOT-L
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| SwinB-DeAOT-Lbackbone=Swin-B, variant=Large, fps=15.42022.10 | 91.1 | — | — | 94.7 | — | — | — | 92.9 | |
| STCNfps=27.22022.10 | 90.8 | — | — | 92.5 | — | — | — | 91.6 | |
| SwinB-AOT-Lbackbone=Swin-B, variant=Large, fps=12.12022.10 | 90.7 | — | — | 93.3 | — | — | — | 92 | |
| R50-DeAOT-Lbackbone=ResNet-50, variant=Large, fps=272022.10 | 90.5 | — | — | 94 | — | — | — | 92.3 | |
| DeAOT-Lvariant=Large, fps=28.52022.10 | 90.3 | — | — | 93.7 | — | — | — | 92 | |
| R50-AOT-Lbackbone=ResNet-50, variant=Large, fps=182022.10 | 90.1 | — | — | 92.1 | — | — | — | 91.1 | |
| HMMNfps=102022.10 | 89.6 | — | — | 92 | — | — | — | 90.8 | |
| AOT-Lvariant=Large, fps=18.72022.10 | 89.6 | — | — | 91.1 | — | — | — | 90.4 | |
| DeAOT-Bvariant=Base, fps=40.92022.10 | 89.4 | — | — | 92.5 | — | — | — | 91 | |
| CFBI+fps=5.92022.10 | 88.7 | — | — | 91.1 | — | — | — | 89.9 | |
| AOT-Bvariant=Base, fps=29.62022.10 | 88.7 | — | — | 91.1 | — | — | — | 89.9 | |
| AOT-Svariant=Small, fps=402022.10 | 88.6 | — | — | 90.2 | — | — | — | 89.4 | |
| DeAOT-Tvariant=Tiny, fps=63.52022.10 | 87.8 | — | — | 89.9 | — | — | — | 88.9 | |
| DeAOT-Svariant=Small, fps=49.22022.10 | 87.6 | — | — | 90.9 | — | — | — | 89.3 | |
| RPCMfps=5.82022.10 | 87.1 | — | — | 94 | — | — | — | 90.6 | |
| AOT-Tvariant=Tiny, fps=51.42022.10 | 86.1 | — | — | 87.4 | — | — | — | 86.8 | |
| MuGSupervision=Weakly-supervised2020.03 | 65.7 | 77.6 | 26.4 | 63.5 | 67.7 | 27.2 | 44.4 | — | |
| MuGSupervision=Unsupervised Learning2020.03 | 63.1 | 71.9 | 28.1 | 61.8 | 64.2 | 30.5 | 43 | — | |
| BVSSupervision=Non Learning2020.03 | 60 | 66.9 | 28.9 | 58.8 | 67.9 | 21.3 | 34.7 | — | |
| FCPSupervision=Non Learning2020.03 | 58.4 | 71.5 | -2 | 49.2 | 49.5 | -1.1 | 30.6 | — | |
| JMPSupervision=Non Learning2020.03 | 57 | 62.6 | 39.4 | 53.1 | 54.2 | 38.4 | 15.9 | — | |
| TimeCycleSupervision=Unsupervised Learning2020.03 | 55.8 | 64.9 | 0 | 51.1 | 51.6 | 2.9 | 36.6 | — | |
| HVSSupervision=Non Learning2020.03 | 54.6 | 61.4 | 23.6 | 52.9 | 61 | 22.7 | 36 | — | |
| SIFT FlowSupervision=Non Learning2020.03 | 51.1 | 58.6 | 18.8 | 44 | 50.3 | 20 | 16.4 | — | |
| CorrFlowSupervision=Unsupervised Learning2020.03 | 48.9 | 44.7 | 19.2 | 39.1 | 28.6 | 17.9 | 36.4 | — | |
| FlowNet2Supervision=Weakly-supervised2020.03 | 41.6 | 45.7 | 19.9 | 40.1 | 38.3 | 26.6 | 29.8 | — | |
| mgPFFSupervision=Unsupervised Learning2020.03 | 40.5 | 34.9 | 18.8 | 34 | 24.2 | 13.8 | 53.1 | — | |
| Vondrick et al.Supervision=Unsupervised Learning2020.03 | 38.9 | 37.1 | 22.4 | 30.8 | 21.7 | 16.7 | 45.9 | — |