No-Frills Human-Object Interaction Detection: Factorization, Layout Encodings, and Training Techniques
About
We show that for human-object interaction detection a relatively simple factorized model with appearance and layout encodings constructed from pre-trained object detectors outperforms more sophisticated approaches. Our model includes factors for detection scores, human and object appearance, and coarse (box-pair configuration) and optionally fine-grained layout (human pose). We also develop training techniques that improve learning efficiency by: (1) eliminating a train-inference mismatch; (2) rejecting easy negatives during mini-batch training; and (3) using a ratio of negatives to positives that is two orders of magnitude larger than existing approaches. We conduct a thorough ablation study to understand the importance of different factors and training techniques using the challenging HICO-Det dataset.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Human-Object Interaction Detection | HICO-DET (test) | mAP (full)20.41 | 493 | |
| Human-Object Interaction Detection | V-COCO (test) | AP (Role, Scenario 1)53.1 | 270 | |
| Human-Object Interaction Detection | HICO-DET | mAP (Full)17.18 | 233 | |
| HOI Detection | HICO-DET (test) | Box mAP (Full)17.18 | 32 | |
| Human-Object Interaction Detection | V-COCO | AP (Role)31.8 | 23 | |
| Human-Object Interaction Detection | HICO-DET 9 (test) | mAP (Full)17.18 | 21 | |
| Human-Object Interaction Detection | HOI-VP | mAP61.05 | 11 |