List your company

Human-Object Interaction Detection on HICO-DET (Rare First Unseen Combination (RF-UC))

44.81mAP (Full)DA-HOI
16.605223.927631.2538.5724Aug 12, 2024Nov 12, 2024Feb 12, 2025May 15, 2025Aug 15, 2025Nov 15, 2025Feb 16, 2026

Evaluation results

MethodLinks
DA-HOIDetector=Grounding-DINO2026.0244.8145.1943.3
DA-HOIDetector=Yolo-World2026.024444.0543.81
DA-HOIDetector=ResNet50 DETR2026.0243.5644.0141.79
GRASP-HOISetting=Zero-shot, Protocol=Open-vocabulary2025.1242.4642.9843.67
BC-HOISetting=Zero-shot, Protocol=Open-vocabulary2025.1240.9940.6742.31
BC-HOIFeatures=BLIP22026.0240.9940.6742.31
SICHOISetting=Zero-shot, Protocol=Open-vocabulary2025.1240.1141.5834.24
LAINBackbone=ViT-L2025.0538.1338.5436.57
CMMPSetting=Zero-shot, Protocol=Open-vocabulary2025.1237.1337.4235.98
EZ-HOIBackbone=Resnet50+ViT-L, Setting=RF2024.1036.7337.3534.24
EZ-HOIBackbone=Resnet50+ViT-L, Setting=RF2024.1036.7337.3534.24
EZ-HOISetting=Zero-shot, Protocol=Open-vocabulary2025.1236.7337.3534.24
EZ-HOI2026.0236.7337.3534.24
FleVRSevaluation_mode=zero-shot2024.0836.639.6126.06
DiffHOISetting=Zero-shot, Protocol=Open-vocabulary2025.1235.8936.7732.06
DP-HOISetting=Zero-shot, Protocol=Open-vocabulary2025.1235.0336.1730.49
LAINBackbone=ViT-B2025.0534.4135.0631.83
LAIN2026.0234.4135.0631.83
CLIP4HOISetting=RF-UC2024.0834.0835.4828.47
CLIP4HOISetting=RF-UC2024.0834.0835.4828.47
CLIP4HOIBackbone=Resnet50+ViT-B, Setting=RF2024.1034.0835.4828.47
CLIP4HOIBackbone=Resnet50+ViT-B, Setting=RF2024.1034.0835.4828.47
CLIP4HOIBackbone=ViT-B2025.0534.0835.4828.47
CLIP4HOI2026.0234.0835.4828.47
HOIGenSetting=RF-UC2024.0833.8634.5731.01
HOIGenSetting=RF-UC2024.0833.8634.5731.01
BCOMBackbone=ViT-L2025.0533.7435.0428.52
BCOMDetector=ResNet50 DETR2026.0233.7435.0428.52
LOGICHOISetting=RF-UC2024.0833.1734.9325.97
LOGICHOISetting=RF-UC2024.0833.1734.9325.97
LOGICHOIBackbone=ViT-B2025.0533.1734.9325.97
EZ-HOIBackbone=Resnet50+ViT-B, Setting=RF2024.1033.1334.1529.02
EZ-HOIBackbone=Resnet50+ViT-B, Setting=RF2024.1033.1334.1529.02
ADA-CMSetting=RF-UC2024.0833.0134.3527.63
ADA-CMSetting=RF-UC2024.0833.0134.3527.63
ADA-CMBackbone=Resnet50+ViT-B, Setting=RF2024.1033.0134.3527.63
ADA-CMBackbone=Resnet50+ViT-B, Setting=RF2024.1033.0134.3527.63
ADA-CMBackbone=ViT-B2025.0533.0134.3527.63
ADA-CMDetector=ResNet50 DETR2026.0233.0134.3527.63
HOICLIPSetting=RF-UC2024.0832.9934.8525.53
HOICLIPSetting=RF-UC2024.0832.9934.8525.53
HOICLIPBackbone=Resnet50+ViT-B, Setting=RF2024.1032.9928.4725.53
HOICLIPBackbone=Resnet50+ViT-B, Setting=RF2024.1032.9928.4725.53
HOICLIPBackbone=ViT-B2025.0532.9934.8525.53
HOICLIPSetting=Zero-shot, Protocol=Open-vocabulary2025.1232.9934.8525.53
HOICLIP2026.0232.9934.8525.53
RLIPv2-ParSeDASetting=Zero-shot, Protocol=Open-vocabulary2025.1232.9735.8521.45
UniHOIBackbone=Resnet50+ViT-L, Setting=RF2024.1032.2733.1628.68
UniHOIBackbone=Resnet50+ViT-L, Setting=RF2024.1032.2733.1628.68
UniHOI (BLIP2)Setting=Zero-shot, Protocol=Open-vocabulary2025.1232.2733.1628.68
UniHOIFeatures=BLIP22026.0232.2733.1628.68
CMMPBackbone=ViT-B2025.0532.1832.8729.45
CMMP2026.0232.1832.8729.45
DA-HOITraining-free=true2026.0231.5——
GEN-VLKTSetting=RF-UC2024.0830.5632.9121.36
GEN-VLKTSetting=RF-UC2024.0830.5632.9121.36
GEN-VLKTevaluation_mode=zero-shot2024.0830.5632.9121.36
GEN-VLKTBackbone=Resnet50+ViT-B, Setting=RF2024.1030.5632.9121.36
GEN-VLKTBackbone=Resnet50+ViT-B, Setting=RF2024.1030.5632.9121.36
GEN-VLKTBackbone=ViT-B2025.0530.5632.9121.36
GEN-VLKTSetting=Zero-shot, Protocol=Open-vocabulary2025.1230.5632.9121.36
GEN-VLKT2026.0230.5632.9121.36
RLIPBackbone=ViT-B2025.0530.5233.3519.19
EoIDBackbone=Resnet50, Setting=RF2024.1029.5231.3922.04
EoIDBackbone=Resnet50, Setting=RF2024.1029.5231.3922.04
EoIDBackbone=ViT-B2025.0529.5231.3922.04
ADA-CMDetector=Grounding-DINO2026.0228.2630.1220.82
ADA-CMDetector=Yolo-World2026.0227.4729.2320.43
ADA-CMTraining-free=true2026.0225.19——
CLIPBackbone=ViT-B2025.0523.362228.79
FCLevaluation_mode=zero-shot2024.0822.0124.2313.16
FCLBackbone=ViT-B2025.0522.0124.2313.16
ATLevaluation_mode=zero-shot2024.0821.5724.679.18
ATLBackbone=ViT-B2025.0521.5724.679.18
VCLevaluation_mode=zero-shot2024.0821.4324.2810.06
BCOMDetector=Grounding-DINO2026.0220.3123.089.23
BCOMDetector=Yolo-World2026.0217.6920.078.17