Person Re‐identification Based on Feature Erasure and Diverse Feature Learning
Abstract Robust and comprehensive modelling of targets is the key to successful person re‐identification. However, some useful information may be ignored, since CNNs tend to learn from the most distinctive feature region of the human body. In the present study, a multi‐branch lightweight network str...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-09-01
|
| Series: | IET Computer Vision |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/cvi2.12108 |
| _version_ | 1811345105494212608 |
|---|---|
| author | Lu Peng Zhang Jidong Zhang Zhen Wang Wei Dou Yamei |
| author_facet | Lu Peng Zhang Jidong Zhang Zhen Wang Wei Dou Yamei |
| author_sort | Lu Peng |
| collection | DOAJ |
| description | Abstract Robust and comprehensive modelling of targets is the key to successful person re‐identification. However, some useful information may be ignored, since CNNs tend to learn from the most distinctive feature region of the human body. In the present study, a multi‐branch lightweight network structure that can enhance the ability of diverse feature retrieval is introduced. The proposed network consists of three branches. In the feature erasure branch, a drop block model is added to remove the horizontal region with the highest activation degree from feature vectors so as to allow the network to learn relatively low discrimination features. The global branch is used as an essential supplement to the feature erasure branch. A unified horizontal segmentation strategy is adopted in the local branch to avoid the influence of feature dislocation. Finally, diverse feature learning is achieved through the branch network structure. The proposed method can achieve state‐of‐the‐art results on Market‐1501, CUHK03 and DukeMTMC‐Reid data sets, thereby demonstrating the effectiveness of the method. |
| first_indexed | 2024-04-13T19:58:04Z |
| format | Article |
| id | doaj.art-27587d78a896413c9ba6c3f1b0927452 |
| institution | Directory Open Access Journal |
| issn | 1751-9632 1751-9640 |
| language | English |
| last_indexed | 2024-04-13T19:58:04Z |
| publishDate | 2022-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Computer Vision |
| spelling | doaj.art-27587d78a896413c9ba6c3f1b09274522022-12-22T02:32:16ZengWileyIET Computer Vision1751-96321751-96402022-09-0116650451310.1049/cvi2.12108Person Re‐identification Based on Feature Erasure and Diverse Feature LearningLu Peng0Zhang Jidong1Zhang Zhen2Wang Wei3Dou Yamei4College of Electrical Engineering ZhengZhou University Zhengzhou Henan ChinaCollege of Electrical Engineering ZhengZhou University Zhengzhou Henan ChinaCollege of Electrical Engineering ZhengZhou University Zhengzhou Henan ChinaCollege of Electrical Engineering ZhengZhou University Zhengzhou Henan ChinaCollege of Electrical Engineering ZhengZhou University Zhengzhou Henan ChinaAbstract Robust and comprehensive modelling of targets is the key to successful person re‐identification. However, some useful information may be ignored, since CNNs tend to learn from the most distinctive feature region of the human body. In the present study, a multi‐branch lightweight network structure that can enhance the ability of diverse feature retrieval is introduced. The proposed network consists of three branches. In the feature erasure branch, a drop block model is added to remove the horizontal region with the highest activation degree from feature vectors so as to allow the network to learn relatively low discrimination features. The global branch is used as an essential supplement to the feature erasure branch. A unified horizontal segmentation strategy is adopted in the local branch to avoid the influence of feature dislocation. Finally, diverse feature learning is achieved through the branch network structure. The proposed method can achieve state‐of‐the‐art results on Market‐1501, CUHK03 and DukeMTMC‐Reid data sets, thereby demonstrating the effectiveness of the method.https://doi.org/10.1049/cvi2.12108image recognitionimage retrievallearning (artificial intelligence) |
| spellingShingle | Lu Peng Zhang Jidong Zhang Zhen Wang Wei Dou Yamei Person Re‐identification Based on Feature Erasure and Diverse Feature Learning IET Computer Vision image recognition image retrieval learning (artificial intelligence) |
| title | Person Re‐identification Based on Feature Erasure and Diverse Feature Learning |
| title_full | Person Re‐identification Based on Feature Erasure and Diverse Feature Learning |
| title_fullStr | Person Re‐identification Based on Feature Erasure and Diverse Feature Learning |
| title_full_unstemmed | Person Re‐identification Based on Feature Erasure and Diverse Feature Learning |
| title_short | Person Re‐identification Based on Feature Erasure and Diverse Feature Learning |
| title_sort | person re identification based on feature erasure and diverse feature learning |
| topic | image recognition image retrieval learning (artificial intelligence) |
| url | https://doi.org/10.1049/cvi2.12108 |
| work_keys_str_mv | AT lupeng personreidentificationbasedonfeatureerasureanddiversefeaturelearning AT zhangjidong personreidentificationbasedonfeatureerasureanddiversefeaturelearning AT zhangzhen personreidentificationbasedonfeatureerasureanddiversefeaturelearning AT wangwei personreidentificationbasedonfeatureerasureanddiversefeaturelearning AT douyamei personreidentificationbasedonfeatureerasureanddiversefeaturelearning |