A Solution to Co-Occurence Bias: Attributes Disentanglement via Mutual Information Minimization for Pedestrian Attribute Recognition
Abstract
Recent studies on pedestrian attribute recognition progress with either explicit or implicit modeling of the co-occurence among attributes. Considering that this known a prior is highly variable and unforeseeable regarding the specific scenarios, we show that current methods can actually suffer in generalizing such fitted attributes interdependencies onto scenes or identities off the dataset distribution, resulting in the underlined bias of attributes co-occurence. To render models robust in realistic scenes, we propose the attributes-disentangled feature learning to ensure the recognition of an attribute not inferring on the existence of others, and which is sequentially formulated as a problem of mutual information minimization. Rooting from it, practical strategies are devised to efficiently decouple attributes, which substantially improve the baseline and establish state-of-the-art performance on realistic datasets like PETAzs and RAPzs.
Cite
Text
Zhou et al. "A Solution to Co-Occurence Bias: Attributes Disentanglement via Mutual Information Minimization for Pedestrian Attribute Recognition." International Joint Conference on Artificial Intelligence, 2023. doi:10.24963/IJCAI.2023/203Markdown
[Zhou et al. "A Solution to Co-Occurence Bias: Attributes Disentanglement via Mutual Information Minimization for Pedestrian Attribute Recognition." International Joint Conference on Artificial Intelligence, 2023.](https://mlanthology.org/ijcai/2023/zhou2023ijcai-solution/) doi:10.24963/IJCAI.2023/203BibTeX
@inproceedings{zhou2023ijcai-solution,
title = {{A Solution to Co-Occurence Bias: Attributes Disentanglement via Mutual Information Minimization for Pedestrian Attribute Recognition}},
author = {Zhou, Yibo and Hu, Hai-Miao and Yu, Jinzuo and Xu, Zhenbo and Lu, Weiqing and Cao, Yuran},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2023},
pages = {1831-1839},
doi = {10.24963/IJCAI.2023/203},
url = {https://mlanthology.org/ijcai/2023/zhou2023ijcai-solution/}
}