Boundary Detection on NYUD2
78.43ODS FmaxMLORE
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MLOREBackbone=ViT-large2024.03 | 78.43 | — | — | |
| TaskExpertBackbone=ViT-large2024.03 | 78.4 | — | — | |
| TaskPrompterBackbone=ViT-large2024.03 | 78.2 | — | — | |
| InvPTBackbone=ViT-large2024.03 | 78.1 | — | — | |
| ATRCBackbone=HRNet482024.03 | 77.94 | — | — | |
| MTI-NetBackbone=HRNet482024.03 | 77.86 | — | — | |
| PAD-NetBackbone=HRNet182024.03 | 76.38 | — | — | |
| Our(SE+SH + all cues)Input Modality=RGB-D2014.07 | 71.03 | 72.33 | 73.81 | |
| Our(SE + all cues)Input Modality=RGB-D2014.07 | 70.25 | 71.59 | 69.28 | |
| Our(SE + normal gradients)Input Modality=RGB-D2014.07 | 69.55 | 70.89 | 69.32 | |
| SE+SHInput Modality=RGB-D2014.07 | 69.46 | 70.84 | 71.88 | |
| Gupta et al. CVPRInput Modality=RGB-D2014.07 | 68.66 | 71.57 | 62.91 | |
| SEInput Modality=RGB-D2014.07 | 68.45 | 69.92 | 67.93 | |
| Silberman et al.Input Modality=RGB-D2014.07 | 65.77 | 66.06 | — | |
| gPb-ucmInput Modality=RGB2014.07 | 63.15 | 66.12 | 56.2 |