Human-Object Interaction Detection on HICO-DET (NF-UC)
42.01mAP (Full)DA-HOI
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DA-HOIDetector=Yolo-World2026.02 | 42.01 | 43.36 | 41.67 | |
| DA-HOIDetector=Grounding-DINO2026.02 | 41.51 | 41.52 | 41.51 | |
| DA-HOIDetector=ResNet50 DETR2026.02 | 40.33 | 43.12 | 39.63 | |
| InCoM-NetBackbone=ViT-L, Zero-shot=true2026.04 | 38.62 | 39.45 | 38.42 | |
| LAINBackbone=ViT-L2025.05 | 36.81 | 32.25 | 37.95 | |
| VDRPBackbone=ViT-L, Zero-shot=true2026.04 | 36.46 | 37.48 | 36.21 | |
| BC-HOIFeatures=BLIP22026.02 | 36.4 | 33.01 | 37.24 | |
| LAINBackbone=ViT-L2025.05 | 36.22 | 37.52 | 35.9 | |
| LAINBackbone=ViT-L, Zero-shot=true2026.04 | 36.22 | 37.52 | 35.9 | |
| InCoM-NetBackbone=ViT-B, Zero-shot=true2026.04 | 35.74 | 38.68 | 35 | |
| InterProDABackbone=ViT-L, Zero-shot=true2026.04 | 35.5 | 33.64 | 36.47 | |
| CMMPBackbone=ViT-L, Zero-shot=true2026.04 | 35.13 | 33.52 | 35.53 | |
| EZ-HOI2026.02 | 34.84 | 36.33 | 34.47 | |
| LAINBackbone=ViT-B2025.05 | 34.36 | 31.64 | 35.04 | |
| LAINBackbone=ViT-B2025.05 | 33.23 | 36.41 | 32.44 | |
| LAIN2026.02 | 33.23 | 36.41 | 32.44 | |
| LAINBackbone=ViT-B, Zero-shot=true2026.04 | 33.23 | 36.41 | 32.44 | |
| VDRPBackbone=ViT-B, Zero-shot=true2026.04 | 32.57 | 36.45 | 31.6 | |
| HOLaBackbone=ViT-B, Zero-shot=true2026.04 | 32.36 | 35.25 | 31.64 | |
| CLIP4HOIBackbone=ViT-B2025.05 | 32.11 | 27.71 | 33.25 | |
| BCOMBackbone=ViT-L2025.05 | 32.03 | 33.12 | 31.76 | |
| BCOMDetector=ResNet50 DETR2026.02 | 32.03 | 33.12 | 31.76 | |
| BCOMBackbone=ViT-L, Zero-shot=true2026.04 | 32.03 | 33.12 | 31.76 | |
| CMMPBackbone=ViT-B2025.05 | 31.84 | 29.6 | 32.39 | |
| UniHOIFeatures=BLIP22026.02 | 31.79 | 28.45 | 32.63 | |
| UniHOIBackbone=ViT-L, Zero-shot=true2026.04 | 31.79 | 28.45 | 32.63 | |
| DA-HOITraining-free=true2026.02 | 31.5 | — | — | |
| ADA-CMBackbone=ViT-B2025.05 | 31.39 | 32.41 | 31.13 | |
| ADA-CMDetector=ResNet50 DETR2026.02 | 31.39 | 32.41 | 31.13 | |
| ADA-CMBackbone=ViT-B, Zero-shot=true2026.04 | 31.39 | 32.41 | 31.13 | |
| CMMPBackbone=ViT-B2025.05 | 30.18 | 32.09 | 29.71 | |
| CMMP2026.02 | 30.18 | 32.09 | 29.71 | |
| CMMPBackbone=ViT-B, Zero-shot=true2026.04 | 30.18 | 32.09 | 29.71 | |
| HOICLIPBackbone=ViT-B2025.05 | 29.93 | 23.15 | 31.65 | |
| EoIDBackbone=ViT-B2025.05 | 28.91 | 23.01 | 30.39 | |
| CLIP4HOIBackbone=ViT-B2025.05 | 28.9 | 31.44 | 28.26 | |
| CLIP4HOI2026.02 | 28.9 | 31.44 | 28.26 | |
| CLIP4HOIBackbone=ViT-B, Zero-shot=true2026.04 | 28.9 | 31.44 | 28.26 | |
| LOGICHOIBackbone=ViT-B2025.05 | 27.95 | 26.84 | 27.86 | |
| HOICLIPBackbone=ViT-B2025.05 | 27.75 | 26.39 | 28.1 | |
| HOICLIP2026.02 | 27.75 | 26.39 | 28.1 | |
| HOICLIPBackbone=ViT-B, Zero-shot=true2026.04 | 27.75 | 26.39 | 28.1 | |
| ADA-CMDetector=Yolo-World2026.02 | 27.13 | 24.89 | 27.69 | |
| ADA-CMDetector=Grounding-DINO2026.02 | 26.87 | 26.23 | 27.03 | |
| EoIDBackbone=ViT-B2025.05 | 26.69 | 26.77 | 26.66 | |
| RLIPBackbone=ViT-B2025.05 | 26.19 | 20.27 | 27.67 | |
| ADA-CMTraining-free=true2026.02 | 25.19 | — | — | |
| GEN-VLKTBackbone=ViT-B2025.05 | 23.71 | 25.05 | 23.38 | |
| GEN-VLKT2026.02 | 23.71 | 25.05 | 23.38 | |
| GEN-VLKTBackbone=ViT-B, Zero-shot=true2026.04 | 23.71 | 25.05 | 23.38 | |
| CLIPBackbone=ViT-B2025.05 | 23.36 | 28.52 | 22.06 | |
| CLIPBackbone=ViT-B2025.05 | 23.36 | 24.28 | 23.12 | |
| BCOMDetector=Grounding-DINO2026.02 | 19.86 | 16.38 | 20.73 | |
| FCLBackbone=ViT-B2025.05 | 19.37 | 18.66 | 19.55 | |
| ATLBackbone=ViT-B2025.05 | 18.67 | 18.25 | 18.78 | |
| BCOMDetector=Yolo-World2026.02 | 17 | 13.76 | 17.81 |