Video Object Discovery on MOVi-C synthetic (test)
73.2ARIRandSF.Q + ICC
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| RandSF.Q + ICC#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 73.2 | 67.4 | 32.9 | 30.3 | |
| RandSF.Q#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 65.4 | 67.4 | 29.2 | 26.8 | |
| SlotContrast#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 64.6 | 59.9 | 27.7 | 25.8 | |
| RandSF.Q + ECC#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 53.8 | 46.6 | 20.5 | 17.6 | |
| SmoothSA + ICC#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 52.1 | 71.2 | 33.4 | 32.1 | |
| SmoothSA#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 50.9 | 69 | 31.7 | 30.2 | |
| SmoothSA + ECC#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 44.1 | 70.4 | 30.6 | 29.2 | |
| VideoSAUR#slot=11, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 41.9 | 53.3 | 16.1 | 14.8 |