Multi-Stream Refining Network for Person Re-Identification
Viewpoint change, pose variation and background clutter have adverse impacts on similarity evaluation for person re-identification. Because of its distinction and reliability, person saliency has been applied to model person appearance characteristics. However, such valuable information is not fully...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9311208/ |
| _version_ | 1818918739681214464 |
|---|---|
| author | Xu Wang Yan Huang Qicong Wang Yan Chen Yehu Shen |
| author_facet | Xu Wang Yan Huang Qicong Wang Yan Chen Yehu Shen |
| author_sort | Xu Wang |
| collection | DOAJ |
| description | Viewpoint change, pose variation and background clutter have adverse impacts on similarity evaluation for person re-identification. Because of its distinction and reliability, person saliency has been applied to model person appearance characteristics. However, such valuable information is not fully exploited to compute similarities of person images with existing deep methods. To this end, we present a novel multi-stream refining based deep multi-task learning scheme that aggregates multi-stage salient embedding features in the network to boost the retrieval performance. Specifically, the backbone network is divided into four stages and a channel significance self-learning sub-module is introduced to strengthen the importance of saliency channels adaptively. Meanwhile, an enhancement sub-module is employed to extract the common information and different information from the channels. Finally, a multi-stream multi-task learning framework combining four-stage branches is adopted to learn discriminative features. Compared with the state-of-the-art approaches, our model achieves competitive performance on three publicly available datasets, i.e., Market-1501, MSMT17, and CUHK03. The experimental results demonstrate the superiority of our method, which achieves 95.67%/88.51%, 87.53%/65.54%, and 89.32%/78.99% on Rank-1/mAP, respectively. |
| first_indexed | 2024-12-20T00:54:45Z |
| format | Article |
| id | doaj.art-cb4500655da442e9a47fa739c518fc03 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T00:54:45Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-cb4500655da442e9a47fa739c518fc032022-12-21T19:59:09ZengIEEEIEEE Access2169-35362021-01-0196596660710.1109/ACCESS.2020.30481199311208Multi-Stream Refining Network for Person Re-IdentificationXu Wang0https://orcid.org/0000-0002-6853-4354Yan Huang1https://orcid.org/0000-0001-7868-093XQicong Wang2https://orcid.org/0000-0001-7324-0433Yan Chen3https://orcid.org/0000-0003-0409-9485Yehu Shen4https://orcid.org/0000-0002-8917-719XShenzhen Research Institute, Xiamen University, Shenzhen, ChinaShenzhen Research Institute, Xiamen University, Shenzhen, ChinaShenzhen Research Institute, Xiamen University, Shenzhen, ChinaCollege of Business and Management, Xiamen Huaxia University, Xiamen, ChinaSchool of Mechanical Engineering, Suzhou University of Science and Technology, Suzhou, ChinaViewpoint change, pose variation and background clutter have adverse impacts on similarity evaluation for person re-identification. Because of its distinction and reliability, person saliency has been applied to model person appearance characteristics. However, such valuable information is not fully exploited to compute similarities of person images with existing deep methods. To this end, we present a novel multi-stream refining based deep multi-task learning scheme that aggregates multi-stage salient embedding features in the network to boost the retrieval performance. Specifically, the backbone network is divided into four stages and a channel significance self-learning sub-module is introduced to strengthen the importance of saliency channels adaptively. Meanwhile, an enhancement sub-module is employed to extract the common information and different information from the channels. Finally, a multi-stream multi-task learning framework combining four-stage branches is adopted to learn discriminative features. Compared with the state-of-the-art approaches, our model achieves competitive performance on three publicly available datasets, i.e., Market-1501, MSMT17, and CUHK03. The experimental results demonstrate the superiority of our method, which achieves 95.67%/88.51%, 87.53%/65.54%, and 89.32%/78.99% on Rank-1/mAP, respectively.https://ieeexplore.ieee.org/document/9311208/Salient channelsrefining modulemulti-streammulti-task |
| spellingShingle | Xu Wang Yan Huang Qicong Wang Yan Chen Yehu Shen Multi-Stream Refining Network for Person Re-Identification IEEE Access Salient channels refining module multi-stream multi-task |
| title | Multi-Stream Refining Network for Person Re-Identification |
| title_full | Multi-Stream Refining Network for Person Re-Identification |
| title_fullStr | Multi-Stream Refining Network for Person Re-Identification |
| title_full_unstemmed | Multi-Stream Refining Network for Person Re-Identification |
| title_short | Multi-Stream Refining Network for Person Re-Identification |
| title_sort | multi stream refining network for person re identification |
| topic | Salient channels refining module multi-stream multi-task |
| url | https://ieeexplore.ieee.org/document/9311208/ |
| work_keys_str_mv | AT xuwang multistreamrefiningnetworkforpersonreidentification AT yanhuang multistreamrefiningnetworkforpersonreidentification AT qicongwang multistreamrefiningnetworkforpersonreidentification AT yanchen multistreamrefiningnetworkforpersonreidentification AT yehushen multistreamrefiningnetworkforpersonreidentification |