HOI Detection on HICO-DET v1.0 (test, Default & Known Objects Settings)
40.95mAP (Default, Full)UniHOI-l
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| UniHOI-lBackbone=ResNet-101, Vision-Language Pre-trained Model=BLIP2, Architecture=One-stage2023.11 | 40.95 | 40.27 | 41.32 | 43.26 | 43.12 | 43.25 | |
| UniHOI-mBackbone=ResNet-101, Vision-Language Pre-trained Model=BLIP2, Architecture=One-stage2023.11 | 40.74 | 40.03 | 40.95 | 42.96 | 42.86 | 42.98 | |
| UniHOI-sBackbone=ResNet-50, Vision-Language Pre-trained Model=BLIP2, Architecture=One-stage2023.11 | 40.06 | 39.91 | 40.11 | 42.2 | 42.6 | 42.08 | |
| ViPLO-lBackbone=ViT-B/16, Architecture=Two-stage2023.11 | 37.22 | 35.45 | 37.75 | 40.61 | 38.82 | 41.15 | |
| Xie et al. (large)Backbone=ResNet-101, Architecture=One-stage2023.11 | 36.03 | 33.16 | 36.89 | 38.82 | 35.51 | 39.81 | |
| GEN-VLKT-lBackbone=ResNet-101, Architecture=One-stage2023.11 | 34.96 | 31.18 | 36.08 | 38.22 | 34.36 | 39.37 | |
| ViPLO-sBackbone=ViT-B/32, Architecture=Two-stage2023.11 | 34.95 | 33.83 | 35.28 | 38.15 | 36.77 | 38.56 | |
| GEN-VLKT-mBackbone=ResNet-101, Architecture=One-stage2023.11 | 34.78 | 31.5 | 35.77 | 38.07 | 34.94 | 39.01 | |
| HOICLIPBackbone=ResNet-101, Architecture=One-stage2023.11 | 34.69 | 31.12 | 35.74 | 37.61 | 34.47 | 38.54 | |
| GEN-VLKT-sBackbone=ResNet-50, Architecture=One-stage2023.11 | 33.75 | 29.25 | 35.1 | 36.78 | 32.75 | 37.99 | |
| Liu et al.Backbone=ResNet-50, Architecture=One-stage2023.11 | 33.51 | 30.3 | 34.46 | 36.28 | 33.16 | 37.21 | |
| MURENBackbone=ResNet-50, Architecture=One-stage2023.11 | 32.87 | 28.67 | 34.12 | 35.52 | 30.88 | 36.91 | |
| UPTBackbone=ResNet-101, Architecture=Two-stage2023.11 | 32.31 | 28.55 | 33.44 | 35.65 | 31.6 | 36.86 | |
| CDN-LBackbone=ResNet-101, Architecture=One-stage2023.11 | 32.07 | 27.19 | 33.53 | 34.79 | 29.48 | 36.38 | |
| UPTBackbone=ResNet-50, Architecture=Two-stage2023.11 | 31.66 | 25.9 | 33.36 | 35.05 | 29.27 | 36.77 | |
| CDN-SBackbone=ResNet-50, Architecture=One-stage2023.11 | 31.44 | 27.39 | 32.64 | 34.09 | 29.63 | 35.42 | |
| SSRTBackbone=ResNet-101, Architecture=One-stage2023.11 | 31.34 | 24.31 | 33.32 | — | — | — | |
| QPICBackbone=ResNet-101, Architecture=One-stage2023.11 | 29.9 | 23.92 | 31.69 | 32.38 | 26.06 | 34.27 | |
| SCGBackbone=ResNet-50-FPN, Architecture=Two-stage2023.11 | 29.26 | 24.61 | 30.65 | 32.87 | 27.89 | 34.35 | |
| AS-NetBackbone=ResNet-50, Architecture=One-stage2023.11 | 28.87 | 24.25 | 30.25 | 31.74 | 27.07 | 33.14 | |
| HOI-TransBackbone=ResNet-101, Architecture=One-stage2023.11 | 26.61 | 19.15 | 28.84 | 29.13 | 20.98 | 31.57 | |
| FCLBackbone=ResNet-50, Architecture=Two-stage2023.11 | 25.27 | 20.57 | 26.67 | 27.71 | 22.34 | 28.93 | |
| IDNBackbone=ResNet-50, Architecture=Two-stage2023.11 | 24.58 | 20.33 | 25.86 | 27.89 | 23.64 | 29.16 | |
| DRGBackbone=ResNet-50-FPN, Architecture=Two-stage2023.11 | 24.53 | 19.47 | 26.04 | 27.98 | 23.11 | 29.43 | |
| ATLBackbone=ResNet-50, Architecture=Two-stage2023.11 | 23.81 | 17.43 | 27.42 | 27.38 | 22.09 | 28.96 | |
| VCLBackbone=ResNet-50, Architecture=Two-stage2023.11 | 23.63 | 17.21 | 25.55 | 25.98 | 19.12 | 28.03 | |
| PPDMBackbone=Hourglass-104, Architecture=One-stage2023.11 | 21.94 | 13.97 | 24.32 | 24.81 | 17.09 | 27.12 | |
| DJ-RNBackbone=ResNet-50, Architecture=Two-stage2023.11 | 21.34 | 18.53 | 22.18 | 23.69 | 20.64 | 24.6 | |
| VSGNetBackbone=ResNet-152, Architecture=Two-stage2023.11 | 19.8 | 16.05 | 20.91 | — | — | — |