Human-Object Interaction Detection on HICO-DET RF-UC (test)
37.69Unseen mAPInCoM-Net
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| InCoM-NetBackbone=ViT-L, Zero-shot=true2026.04 | 37.69 | 41.46 | 40.71 | |
| VDRPBackbone=ViT-L, Zero-shot=true2026.04 | 36.72 | 38.48 | 38.13 | |
| LAINBackbone=ViT-L, Zero-shot=true2026.04 | 36.57 | 38.54 | 38.13 | |
| InterProDABackbone=ViT-L, Zero-shot=true2026.04 | 36.38 | 40.88 | 39.58 | |
| CMMPBackbone=ViT-L, Zero-shot=true2026.04 | 35.98 | 37.42 | 37.13 | |
| InCoM-NetBackbone=ViT-B, Zero-shot=true2026.04 | 32.65 | 38.69 | 37.48 | |
| DIFFUSIONHOIType=RF-UC2024.10 | 32.06 | 36.77 | 35.89 | |
| LAINBackbone=ViT-B, Zero-shot=true2026.04 | 31.83 | 35.06 | 34.41 | |
| VDRPBackbone=ViT-B, Zero-shot=true2026.04 | 31.29 | 34.41 | 33.78 | |
| HOLaBackbone=ViT-B, Zero-shot=true2026.04 | 30.61 | 35.08 | 34.19 | |
| CMMPBackbone=ViT-B, Zero-shot=true2026.04 | 29.45 | 32.87 | 32.18 | |
| UniHOIBackbone=ViT-L, Zero-shot=true2026.04 | 28.68 | 33.16 | 32.27 | |
| BCOMBackbone=ViT-L, Zero-shot=true2026.04 | 28.52 | 35.04 | 33.74 | |
| CLIP4HOIBackbone=ViT-B, Zero-shot=true2026.04 | 28.47 | 35.48 | 34.08 | |
| ADA-CMBackbone=ViT-B, Zero-shot=true2026.04 | 27.63 | 34.35 | 33.01 | |
| HOICLIPType=RF-UC2024.10 | 25.53 | 34.85 | 32.99 | |
| HOICLIPBackbone=ViT-B, Zero-shot=true2026.04 | 25.53 | 34.85 | 32.99 | |
| OpenCatType=RF-UC2024.10 | 21.46 | 33.86 | 31.38 | |
| GEN-VLKTType=RF-UC2024.10 | 21.36 | 32.91 | 30.56 | |
| GEN-VLKTBackbone=ViT-B, Zero-shot=true2026.04 | 21.36 | 32.91 | 30.56 | |
| SCLType=RF-UC2024.10 | 19.07 | 30.39 | 28.08 | |
| FCLType=RF-UC2024.10 | 13.16 | 24.23 | 22.01 | |
| ATLType=RF-UC2024.10 | 9.18 | 24.67 | 21.57 |