Surface Normal Estimation on NYU (DSINE)
15.6Mean Angular ErrorReChannel-9B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ReChannel-9BCategory=RGB-native field readout, Backbone=FLUX-Klein 9B2026.07 | 15.6 | 61.2 | |
| Edit2PercCategory=Generative dense prediction methods2026.07 | 15.7 | 61.6 | |
| ReChannel-4BCategory=RGB-native field readout, Backbone=FLUX-Klein 4B2026.07 | 15.8 | 60.8 | |
| MarigoldCategory=Generative dense prediction methods2026.07 | 16.1 | 60.5 | |
| Lotus-DCategory=Generative dense prediction methods2026.07 | 16.2 | 59.8 | |
| DSINECategory=Specialized / discriminative methods2026.07 | 16.4 | 59.6 | |
| GeoWizardCategory=Generative dense prediction methods2026.07 | 17 | 56.5 | |
| Omnidata v2Category=Specialized / discriminative methods2026.07 | 17.2 | 55.5 | |
| GenPerceptCategory=Generative dense prediction methods2026.07 | 18.3 | 56 |