Generative Human-Object Interaction Detection via Differentiable Cognitive Steering of Multi-modal LLMs

Benchmark Model Rank Results
human-object-interaction-detection-on-hico-detGRASP-HOI–mAP: 48.02
human-object-interaction-detection-on-v-cocoGRASP-HOI–AP(S1): 72.5AP(S2): 76.2