Feature-Level Camera Style Transfer for Person Re-Identification
The person re-identification (re-ID) problem has attracted growing interest in the computer vision community. Most public re-ID datasets are captured by multiple non-overlapping cameras, and the same person may appear dissimilar in different camera views due to variances of illuminations, viewpoints...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-07-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/14/7286 |
| _version_ | 1827598034850021376 |
|---|---|
| author | Yang Liu Hao Sheng Shuai Wang Yubin Wu Zhang Xiong |
| author_facet | Yang Liu Hao Sheng Shuai Wang Yubin Wu Zhang Xiong |
| author_sort | Yang Liu |
| collection | DOAJ |
| description | The person re-identification (re-ID) problem has attracted growing interest in the computer vision community. Most public re-ID datasets are captured by multiple non-overlapping cameras, and the same person may appear dissimilar in different camera views due to variances of illuminations, viewpoints and postures. These differences, collectively referred to as camera style variance, make person re-ID still a challenging problem. Recently, researchers have attempted to solve this problem using generative models. The generative adversarial network (GAN) is widely used for the pose transfer or data augmentation to bridge the camera style gap. However, these methods, mostly based on image-level GAN, require huge computational power during the training of generative models. Furthermore, the training process of GAN is separated from the re-ID model, which makes it hard to achieve a global optimal for both models simultaneously. In this paper, the authors propose to alleviate camera style variance in the re-ID problem by adopting a feature-level Camera Style Transfer (CST) model, which can serve as an intra-class augmentation method and enhance the model robustness against camera style variance. Specifically, the proposed CST method transfers the camera style-related information of input features while preserving the corresponding identity information. Moreover, the training process can be embedded into the re-ID model in an end-to-end manner, which means the proposed approach can be deployed with much less time and memory cost. The proposed approach is verified on several different person re-ID baselines. Extensive experiments show the validity of the proposed CST model and its benefits for re-ID performance on the Market-1501 dataset. |
| first_indexed | 2024-03-09T03:43:49Z |
| format | Article |
| id | doaj.art-cb9c8cd9f79e406885e8a52d0ed8aaa0 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T03:43:49Z |
| publishDate | 2022-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-cb9c8cd9f79e406885e8a52d0ed8aaa02023-12-03T14:37:38ZengMDPI AGApplied Sciences2076-34172022-07-011214728610.3390/app12147286Feature-Level Camera Style Transfer for Person Re-IdentificationYang Liu0Hao Sheng1Shuai Wang2Yubin Wu3Zhang Xiong4State Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, Beijing 100191, ChinaState Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, Beijing 100191, ChinaState Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, Beijing 100191, ChinaState Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, Beijing 100191, ChinaState Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, Beijing 100191, ChinaThe person re-identification (re-ID) problem has attracted growing interest in the computer vision community. Most public re-ID datasets are captured by multiple non-overlapping cameras, and the same person may appear dissimilar in different camera views due to variances of illuminations, viewpoints and postures. These differences, collectively referred to as camera style variance, make person re-ID still a challenging problem. Recently, researchers have attempted to solve this problem using generative models. The generative adversarial network (GAN) is widely used for the pose transfer or data augmentation to bridge the camera style gap. However, these methods, mostly based on image-level GAN, require huge computational power during the training of generative models. Furthermore, the training process of GAN is separated from the re-ID model, which makes it hard to achieve a global optimal for both models simultaneously. In this paper, the authors propose to alleviate camera style variance in the re-ID problem by adopting a feature-level Camera Style Transfer (CST) model, which can serve as an intra-class augmentation method and enhance the model robustness against camera style variance. Specifically, the proposed CST method transfers the camera style-related information of input features while preserving the corresponding identity information. Moreover, the training process can be embedded into the re-ID model in an end-to-end manner, which means the proposed approach can be deployed with much less time and memory cost. The proposed approach is verified on several different person re-ID baselines. Extensive experiments show the validity of the proposed CST model and its benefits for re-ID performance on the Market-1501 dataset.https://www.mdpi.com/2076-3417/12/14/7286person re-identificationcamera style transferfeature generationgenerative adversarial networkdata augmentation |
| spellingShingle | Yang Liu Hao Sheng Shuai Wang Yubin Wu Zhang Xiong Feature-Level Camera Style Transfer for Person Re-Identification Applied Sciences person re-identification camera style transfer feature generation generative adversarial network data augmentation |
| title | Feature-Level Camera Style Transfer for Person Re-Identification |
| title_full | Feature-Level Camera Style Transfer for Person Re-Identification |
| title_fullStr | Feature-Level Camera Style Transfer for Person Re-Identification |
| title_full_unstemmed | Feature-Level Camera Style Transfer for Person Re-Identification |
| title_short | Feature-Level Camera Style Transfer for Person Re-Identification |
| title_sort | feature level camera style transfer for person re identification |
| topic | person re-identification camera style transfer feature generation generative adversarial network data augmentation |
| url | https://www.mdpi.com/2076-3417/12/14/7286 |
| work_keys_str_mv | AT yangliu featurelevelcamerastyletransferforpersonreidentification AT haosheng featurelevelcamerastyletransferforpersonreidentification AT shuaiwang featurelevelcamerastyletransferforpersonreidentification AT yubinwu featurelevelcamerastyletransferforpersonreidentification AT zhangxiong featurelevelcamerastyletransferforpersonreidentification |