Unsupervised Object Segmentation on MOVi-C
68.6FG-ARIDINOSAUR (MLP)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| DINOSAUR (MLP)decoder=MLP2023.12 | 68.6 | 39.1 | — | — | — | |
| SPOT-MEvaluation Protocol=Max(Slot Attention, Decoder Masks), Multi-layer=true, MUFASA=true2026.02 | 67.8 | — | 48.2 | — | 49.2 | |
| DINOSAUR-MEvaluation Protocol=Max(Slot Attention, Decoder Masks), Multi-layer=true, MUFASA=true2026.02 | 66.4 | — | 48.3 | — | 49.2 | |
| SPOTEvaluation Protocol=Max(Slot Attention, Decoder Masks)2026.02 | 57.9 | — | 46.4 | — | 47 | |
| DINOSAURdecoder=autoregressive2023.12 | 55.7 | 42.4 | — | — | — | |
| DINOSAUREvaluation Protocol=Max(Slot Attention, Decoder Masks)2026.02 | 55.7 | — | — | — | 42.4 | |
| SPOTensemble=true2023.12 | 52.3 | 47.3 | 46.7 | — | — | |
| SPOTensemble=false2023.12 | 52.1 | 47 | 46.4 | — | — | |
| SLATE+2023.03 | 52.04 | 38.17 | 36.44 | — | — | |
| LSD2023.12 | 52 | 45.6 | 44.2 | — | — | |
| LSD2023.03 | 51.98 | 45.57 | 44.19 | — | — | |
| SLASH2023.12 | 51.9 | — | 27.7 | — | — | |
| SLATE2023.03 | 49.54 | 39.37 | 37.75 | — | — | |
| SLATE2023.12 | 49.5 | 39.4 | 37.8 | — | — | |
| SLATEEvaluation Protocol=Max(Slot Attention, Decoder Masks)2026.02 | 49.5 | — | 37.8 | — | 39.4 | |
| SA2023.12 | 43.8 | 26.2 | — | — | — | |
| SAEvaluation Protocol=Max(Slot Attention, Decoder Masks)2026.02 | 43.8 | — | — | — | 26.2 | |
| 11-block Mask2023.12 | 42.7 | 19.5 | 18.2 | — | — |