No-Frills Human-Object Interaction Detection: Factorization, Layout Encodings, and Training Techniques
Abstract
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.
Cite
Text
Gupta et al. "No-Frills Human-Object Interaction Detection: Factorization, Layout Encodings, and Training Techniques." Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019. doi:10.1109/ICCV.2019.00977Markdown
[Gupta et al. "No-Frills Human-Object Interaction Detection: Factorization, Layout Encodings, and Training Techniques." Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019.](https://mlanthology.org/iccv/2019/gupta2019iccv-nofrills/) doi:10.1109/ICCV.2019.00977BibTeX
@inproceedings{gupta2019iccv-nofrills,
title = {{No-Frills Human-Object Interaction Detection: Factorization, Layout Encodings, and Training Techniques}},
author = {Gupta, Tanmay and Schwing, Alexander and Hoiem, Derek},
booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
year = {2019},
doi = {10.1109/ICCV.2019.00977},
url = {https://mlanthology.org/iccv/2019/gupta2019iccv-nofrills/}
}