Semantic-Guided Shared Feature Alignment for Occluded Person Re-IDentification

Xuena Ren, Dongming Zhang, Xiuguo Bao
Proceedings of The 12th Asian Conference on Machine Learning, PMLR 129:17-32, 2020.

Abstract

Occluded Person Re-ID is a challenging task under resolved. Instead of extracting features over the entire image which would easily cause mismatching, we propose Semantic-Guided Shared Feature Alignment (SGSFA) method to extract features focusing on the non-occluded parts. SGSFA parses human body regions through Semantic Guided (SG) branch and aligns regions through Spatial Feature Alignment (SFA) branch simultaneously, and gets enriched representations over the regions for Re-ID. Dynamic classification loss of spatial features and their dynamical sequential combinations in the training stage help facilitate feature diversity. During the matching stage, we use only the visible feature shared by probe and gallery with no extra cues. The experiment results show that SGSFA achieves rank-1 of 62.3% and 50.5% respectively for Occluded-DukeMTMC and P-DukeMTMC-reID, surpassing the state-of-the-art by a large margin.

Cite this Paper

BibTeX
@InProceedings{pmlr-v129-ren20a, title = {Semantic-Guided Shared Feature Alignment for Occluded Person Re-IDentification}, author = {Ren, Xuena and Zhang, Dongming and Bao, Xiuguo}, booktitle = {Proceedings of The 12th Asian Conference on Machine Learning}, pages = {17--32}, year = {2020}, editor = {Pan, Sinno Jialin and Sugiyama, Masashi}, volume = {129}, series = {Proceedings of Machine Learning Research}, month = {18--20 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v129/ren20a/ren20a.pdf}, url = {https://proceedings.mlr.press/v129/ren20a.html}, abstract = {Occluded Person Re-ID is a challenging task under resolved. Instead of extracting features over the entire image which would easily cause mismatching, we propose Semantic-Guided Shared Feature Alignment (SGSFA) method to extract features focusing on the non-occluded parts. SGSFA parses human body regions through Semantic Guided (SG) branch and aligns regions through Spatial Feature Alignment (SFA) branch simultaneously, and gets enriched representations over the regions for Re-ID. Dynamic classification loss of spatial features and their dynamical sequential combinations in the training stage help facilitate feature diversity. During the matching stage, we use only the visible feature shared by probe and gallery with no extra cues. The experiment results show that SGSFA achieves rank-1 of 62.3% and 50.5% respectively for Occluded-DukeMTMC and P-DukeMTMC-reID, surpassing the state-of-the-art by a large margin. } }
Endnote
%0 Conference Paper %T Semantic-Guided Shared Feature Alignment for Occluded Person Re-IDentification %A Xuena Ren %A Dongming Zhang %A Xiuguo Bao %B Proceedings of The 12th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Sinno Jialin Pan %E Masashi Sugiyama %F pmlr-v129-ren20a %I PMLR %P 17--32 %U https://proceedings.mlr.press/v129/ren20a.html %V 129 %X Occluded Person Re-ID is a challenging task under resolved. Instead of extracting features over the entire image which would easily cause mismatching, we propose Semantic-Guided Shared Feature Alignment (SGSFA) method to extract features focusing on the non-occluded parts. SGSFA parses human body regions through Semantic Guided (SG) branch and aligns regions through Spatial Feature Alignment (SFA) branch simultaneously, and gets enriched representations over the regions for Re-ID. Dynamic classification loss of spatial features and their dynamical sequential combinations in the training stage help facilitate feature diversity. During the matching stage, we use only the visible feature shared by probe and gallery with no extra cues. The experiment results show that SGSFA achieves rank-1 of 62.3% and 50.5% respectively for Occluded-DukeMTMC and P-DukeMTMC-reID, surpassing the state-of-the-art by a large margin.
APA
Ren, X., Zhang, D. & Bao, X.. (2020). Semantic-Guided Shared Feature Alignment for Occluded Person Re-IDentification. Proceedings of The 12th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 129:17-32 Available from https://proceedings.mlr.press/v129/ren20a.html.

Related Material