Referring Video Object Segmentation on A2D-S (test)
72.3oIoUReferFormer
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ReferFormerBackbone=Video-Swin-T, FPS=6.542022.03 | 72.3 | — | 64.1 | 76 | 72.2 | 65.4 | 49.8 | 17.9 | 48.6 | |
| MTTRBackbone=Video-Swin-T, FPS=6.692022.03 | 72 | — | 64 | 75.4 | 71.2 | 63.8 | 48.5 | 16.9 | 46.1 | |
| LOCATERBackbone=Video-Swin-T, FPS=7.922022.03 | 71.6 | — | 64.2 | 76.5 | 72.8 | 66.3 | 49.5 | 17.1 | 48.8 | |
| LOCATERBackbone=ViT-B, FPS=10.12022.03 | 69 | — | 59.7 | 70.9 | 64 | 52.5 | 35.1 | 10.1 | 42.8 | |
| Yang et al.Backbone=ResNet-1012022.03 | 67.9 | — | 52.9 | 61.1 | 55.9 | 48.6 | 34.2 | 12 | — | |
| Hui et al.Backbone=I3D, FPS=8.072022.03 | 66.2 | — | 56.1 | 65.4 | 58.9 | 49.7 | 33.3 | 9.1 | 39.9 | |
| Ning et al.Backbone=I3D, FPS=5.422022.03 | 66.1 | — | 52.9 | 63.4 | 57.9 | 48.3 | 32.2 | 8.3 | 38.8 | |
| Wang et al.Backbone=I3D, FPS=7.182022.03 | 62.3 | — | 53.1 | 60.7 | 52.5 | 40.5 | 23.5 | 4.5 | 33.3 | |
| Wang et al.Backbone=I3D, FPS=8.642022.03 | 60.1 | — | 49 | 55.7 | 45.9 | 31.9 | 16 | 2 | 27.4 | |
| McIntosh et al.Backbone=I3D2022.03 | 56.8 | — | 46 | 52.6 | 45 | 34.5 | 20.7 | 3.6 | 30.3 | |
| Gavrilyuk et al.Backbone=I3D2022.03 | 55.1 | — | 42.6 | 50 | 37.6 | 23.1 | 9.4 | 0.4 | 21.5 | |
| Li et al.Backbone=VGG-162022.03 | 51.5 | — | 35.4 | 38.7 | 29 | 17.5 | 6.6 | 0.1 | 16.3 | |
| Hu et al.Backbone=VGG-162022.03 | 47.4 | — | 35 | 34.8 | 23.6 | 13.3 | 3.3 | 0.1 | 13.2 | |
| CSTM2022.03 | 34.1 | 16.3 | 28.7 | — | — | — | — | — | — | |
| LOCATERMemory=True2022.03 | 31.4 | 19 | 29.9 | — | — | — | — | — | — | |
| LOCATERMemory=False2022.03 | 30 | 17.8 | 28.4 | — | — | — | — | — | — | |
| ACGA2022.03 | 25.3 | 9.7 | 20.1 | — | — | — | — | — | — |