Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-Identification
Visible-infrared person re-identification aims to solve the matching problem between cross-camera and cross-modal person images. Existing methods strive to perform better cross-modal alignment, but often neglect the critical importance of feature enhancement for achieving better performance. Therefo...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/11/4988 |
| _version_ | 1827739231014879232 |
|---|---|
| author | Ronghui Lin Rong Wang Wenjing Zhang Ao Wu Yihan Bi |
| author_facet | Ronghui Lin Rong Wang Wenjing Zhang Ao Wu Yihan Bi |
| author_sort | Ronghui Lin |
| collection | DOAJ |
| description | Visible-infrared person re-identification aims to solve the matching problem between cross-camera and cross-modal person images. Existing methods strive to perform better cross-modal alignment, but often neglect the critical importance of feature enhancement for achieving better performance. Therefore, we proposed an effective method that combines both modal alignment and feature enhancement. Specifically, we introduced Visible-Infrared Modal Data Augmentation (VIMDA) for visible images to improve modal alignment. Margin MMD-ID Loss was also used to further enhance modal alignment and optimize model convergence. Then, we proposed Multi-Grain Feature Extraction (MGFE) Structure for feature enhancement to further improve recognition performance. Extensive experiments have been carried out on SYSY-MM01 and RegDB. The result indicates that our method outperforms the current state-of-the-art method for visible-infrared person re-identification. Ablation experiments verified the effectiveness of the proposed method. |
| first_indexed | 2024-03-11T02:57:54Z |
| format | Article |
| id | doaj.art-a3ec7a831c86473ead83b589c3933e46 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T02:57:54Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-a3ec7a831c86473ead83b589c3933e462023-11-18T08:31:00ZengMDPI AGSensors1424-82202023-05-012311498810.3390/s23114988Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-IdentificationRonghui Lin0Rong Wang1Wenjing Zhang2Ao Wu3Yihan Bi4School of Information and Cyber Security, People’s Public Security University of China, Beijing 100038, ChinaSchool of Information and Cyber Security, People’s Public Security University of China, Beijing 100038, ChinaSchool of Information and Cyber Security, People’s Public Security University of China, Beijing 100038, ChinaSchool of Information and Cyber Security, People’s Public Security University of China, Beijing 100038, ChinaSchool of Information and Cyber Security, People’s Public Security University of China, Beijing 100038, ChinaVisible-infrared person re-identification aims to solve the matching problem between cross-camera and cross-modal person images. Existing methods strive to perform better cross-modal alignment, but often neglect the critical importance of feature enhancement for achieving better performance. Therefore, we proposed an effective method that combines both modal alignment and feature enhancement. Specifically, we introduced Visible-Infrared Modal Data Augmentation (VIMDA) for visible images to improve modal alignment. Margin MMD-ID Loss was also used to further enhance modal alignment and optimize model convergence. Then, we proposed Multi-Grain Feature Extraction (MGFE) Structure for feature enhancement to further improve recognition performance. Extensive experiments have been carried out on SYSY-MM01 and RegDB. The result indicates that our method outperforms the current state-of-the-art method for visible-infrared person re-identification. Ablation experiments verified the effectiveness of the proposed method.https://www.mdpi.com/1424-8220/23/11/4988person re-identificationvisible-infrared imagesmodal alignmentmulti-graindata augmentation |
| spellingShingle | Ronghui Lin Rong Wang Wenjing Zhang Ao Wu Yihan Bi Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-Identification Sensors person re-identification visible-infrared images modal alignment multi-grain data augmentation |
| title | Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-Identification |
| title_full | Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-Identification |
| title_fullStr | Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-Identification |
| title_full_unstemmed | Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-Identification |
| title_short | Joint Modal Alignment and Feature Enhancement for Visible-Infrared Person Re-Identification |
| title_sort | joint modal alignment and feature enhancement for visible infrared person re identification |
| topic | person re-identification visible-infrared images modal alignment multi-grain data augmentation |
| url | https://www.mdpi.com/1424-8220/23/11/4988 |
| work_keys_str_mv | AT ronghuilin jointmodalalignmentandfeatureenhancementforvisibleinfraredpersonreidentification AT rongwang jointmodalalignmentandfeatureenhancementforvisibleinfraredpersonreidentification AT wenjingzhang jointmodalalignmentandfeatureenhancementforvisibleinfraredpersonreidentification AT aowu jointmodalalignmentandfeatureenhancementforvisibleinfraredpersonreidentification AT yihanbi jointmodalalignmentandfeatureenhancementforvisibleinfraredpersonreidentification |