Object-Centric Representation Learning on VOC
58.9mBOiGLASS
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GLASSCategory=Diffusion2026.03 | 58.9 | 62.2 | 58.1 | |
| SlotDiffusionCategory=Diffusion2026.03 | 50.4 | 55.3 | — | |
| SPOT+FS+RCCategory=Autoregressive Teacher-Student2026.03 | 49.3 | 56.5 | — | |
| SPOTCategory=Autoregressive Teacher-Student2026.03 | 48.3 | 55.6 | 46.8 | |
| Slot Merging2026.03 | 47.83 | 55.34 | 46.01 | |
| Slot MergingBackbone=ViT-B/16, Resolution=224x2242026.03 | 47.83 | 55.34 | — | |
| DINOSAUR+FS+RCCategory=DINOSAUR-Based2026.03 | 45.2 | 52.9 | — | |
| DINOSAUR+FS+RC2026.03 | 45.2 | 52.9 | — | |
| DIAS_image2026.03 | 44.8 | — | 42.8 | |
| DIASimage2026.03 | 44.8 | — | — | |
| DINOSAURCategory=DINOSAUR-Based2026.03 | 44 | 51.2 | — | |
| DINOSAUR2026.03 | 44 | 51.2 | — | |
| Top-Down Slot Attention2026.03 | 43.9 | 51 | 42 | |
| MetaSlot-MLPCategory=Adaptive Slot Attention2026.03 | 43.9 | — | 42.1 | |
| MetaSlot-MLP2026.03 | 43.9 | — | — | |
| Rotating Features2026.03 | 40.7 | 46 | — | |
| Rotating Features2026.03 | 40.7 | 46 | — | |
| DINOSAUR-MLPCategory=DINOSAUR-Based2026.03 | 39.5 | 40.9 | — | |
| DINOSAUR-MLP2026.03 | 39.5 | 40.9 | — | |
| MetaSlot-TfdCategory=Adaptive Slot Attention2026.03 | 39 | — | 37.8 | |
| MetaSlot-Tfd2026.03 | 39 | — | — | |
| MetaSlot-DfzCategory=Adaptive Slot Attention2026.03 | 36.5 | — | 35.3 | |
| SLATE2026.03 | 35.9 | 41.5 | — | |
| SLATE2026.03 | 35.9 | 41.5 | — | |
| CAE2026.03 | 32.9 | 37.4 | — | |
| CAE2026.03 | 32.9 | 37.4 | — | |
| SA2026.03 | 24.6 | 24.9 | — | |
| SA2026.03 | 24.6 | 24.9 | — |