Object Recognition on YTVIS 2022
90.7Class Top-1 AccuracyRandSF.Q
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| RandSF.QNumber of slots=7, Evaluation protocol=+MLP, Backbone=DINO2 ViT-S/14, Input resolution=256×256 (224×224)2025.08 | 90.7 | 97.1 | 46.5 | 19,120 | |
| SlotContrastNumber of slots=7, Evaluation protocol=+MLP, Backbone=DINO2 ViT-S/14, Input resolution=256×256 (224×224)2025.08 | 87.1 | 96.4 | 48.2 | 19,943 |