Video Object Discovery on MOVi-E synthetic (test)
41.6ARIRandSF.Q + ICC
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| RandSF.Q + ICC#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 41.6 | 80.5 | 26.3 | 24.8 | |
| SmoothSA#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 36.7 | 73.6 | 28.6 | 27.4 | |
| SmoothSA + ICC#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 35.4 | 74 | 28.7 | 27.5 | |
| SmoothSA + ECC#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 35 | 65.2 | 23.9 | 22.6 | |
| RandSF.Q + ECC#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 34 | 45.1 | 13.3 | 11.8 | |
| RandSF.Q#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 30.5 | 82.1 | 23 | 21.6 | |
| SlotContrast#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 29.9 | 70.6 | 20.7 | 19.3 | |
| VideoSAUR#slot=21, conditional=true, Input resolution=224x224, Encoder=DINO2 ViT-S/14, Seeds=42, 43, 442026.05 | 17.4 | 34.6 | 8.3 | 7.5 |