Feature Erasing and Diffusion Network for Occluded Person Re-Identification

Zhikang Wang, Feng Zhu, Shixiang Tang, Rui Zhao, Lihuo He, Jiangning Song; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 4754-4763

Abstract


Occluded person re-identification (ReID) aims at matching occluded person images to holistic ones across different camera views. Target Pedestrians (TP) are often disturbed by Non-Pedestrian Occlusions (NPO) and Non-Target Pedestrians (NTP). Previous methods mainly focus on increasing the model's robustness against NPO while ignoring feature contamination from NTP. In this paper, we propose a novel Feature Erasing and Diffusion Network (FED) to simultaneously handle challenges from NPO and NTP. Specifically, aided by the NPO augmentation strategy that simulates NPO on holistic pedestrian images and generates precise occlusion masks, NPO features are explicitly eliminated by our proposed Occlusion Erasing Module (OEM). Subsequently, we diffuse the pedestrian representations with other memorized features to synthesize the NTP characteristics in the feature space through the novel Feature Diffusion Module (FDM). With the guidance of the occlusion scores from OEM, the feature diffusion process is conducted on visible body parts, thereby improving the quality of the synthesized NTP characteristics. We can greatly improve the model's perception ability towards TP and alleviate the influence of NPO and NTP by jointly optimizing OEM and FDM. Furthermore, the proposed FDM works as an auxiliary module for training and will not be engaged in the inference phase, thus with high flexibility. Experiments on occluded and holistic person ReID benchmarks demonstrate the superiority of FED over state-of-the-art methods.

Related Material


[pdf] [arXiv]
[bibtex]
@InProceedings{Wang_2022_CVPR, author = {Wang, Zhikang and Zhu, Feng and Tang, Shixiang and Zhao, Rui and He, Lihuo and Song, Jiangning}, title = {Feature Erasing and Diffusion Network for Occluded Person Re-Identification}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2022}, pages = {4754-4763} }