Interactive Mask Correction on DAVIS From Initial Mask 585
1.8NoC 85MST
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MSTBackbone=ViT-B2024.01 | 1.8 | 2.29 | — | 22 | — | — | |
| SimpleClickBackbone=ViT-L2024.01 | 1.81 | 2.57 | — | 25 | — | — | |
| FocalClick-segformerB3-S2Backbone=SegFormer-B3, Input Resolution=S2 (256x256), Training Data=COCO+LVIS2022.04 | 2 | 2.76 | 4.3 | 22 | 35 | 53 | |
| FocalClickBackbone=B3-S22024.01 | 2 | 2.76 | — | 22 | — | — | |
| FocalClick-segformerB0-S2Backbone=SegFormer-B0, Input Resolution=S2 (256x256), Training Data=COCO+LVIS2022.04 | 2.2 | 3.08 | 4.82 | 27 | 39 | 68 | |
| SimpleClickBackbone=ViT-B2024.01 | 2.24 | 3.1 | — | 25 | — | — | |
| FocalClick-hrnet32-S2Backbone=HRNet-32, Input Resolution=S2 (256x256), Training Data=COCO+LVIS2022.04 | 2.32 | 3.09 | 4.94 | 28 | 41 | 74 | |
| FocalClickBackbone=hrnet32-S22024.01 | 2.32 | 3.09 | — | 28 | — | — | |
| FocalClick-hrnet18s-S2Backbone=HRNet-18s, Input Resolution=S2 (256x256), Training Data=COCO+LVIS2022.04 | 2.48 | 3.34 | 5.18 | 31 | 43 | 79 | |
| FocalClick-segformerB0-S1Backbone=SegFormer-B0, Input Resolution=S1 (128x128), Training Data=COCO+LVIS2022.04 | 2.63 | 3.69 | 6.08 | 38 | 54 | 104 | |
| FocalClick-hrnet18s-S1Backbone=HRNet-18s, Input Resolution=S1 (128x128), Training Data=COCO+LVIS2022.04 | 2.72 | 3.82 | 5.86 | 37 | 57 | 97 | |
| SimpleClickBackbone=ViT-L2024.01 | 3.39 | 4.88 | — | 36 | — | — | |
| RITM-hrnet32Backbone=HRNet-32, Training Data=COCO+LVIS2022.04 | 3.68 | 5.57 | 11.35 | 46 | 75 | 214 | |
| RITMBackbone=hrnet322024.01 | 3.68 | 5.57 | — | 46 | — | — | |
| RITM-hrnet18sBackbone=HRNet-18s, Training Data=COCO+LVIS2022.04 | 3.71 | 5.96 | 11.83 | 49 | 80 | 235 | |
| RITMBackbone=hrnet18s2024.01 | 3.71 | 5.96 | — | 49 | — | — | |
| MSTBackbone=ViT-B2024.01 | 3.78 | 5.22 | — | 43 | — | — | |
| SimpleClickBackbone=ViT-B2024.01 | 4.06 | 5.83 | — | 42 | — | — | |
| FocalClickBackbone=B3-S22024.01 | 4.06 | 5.89 | — | 43 | — | — | |
| RITMBackbone=hrnet322024.01 | 4.74 | 6.74 | — | 45 | — | — | |
| FocalClickBackbone=hrnet32-S22024.01 | 4.77 | 6.84 | — | 48 | — | — | |
| RITMBackbone=hrnet18s2024.01 | 5.34 | 7.57 | — | 52 | — | — |