Impostor Resilient Multimodal Metric Learning for Person Reidentification
In person reidentification distance metric learning suffers a great challenge from impostor persons. Mostly, distance metrics are learned by maximizing the similarity between positive pair against impostors that lie on different transform modals. In addition, these impostors are obtained from Galler...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2018-01-01
|
| Series: | Advances in Multimedia |
| Online Access: | http://dx.doi.org/10.1155/2018/3202495 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832562839760404480 |
|---|---|
| author | Muhamamd Adnan Syed Zhenjun Han Zhaoju Li Jianbin Jiao |
| author_facet | Muhamamd Adnan Syed Zhenjun Han Zhaoju Li Jianbin Jiao |
| author_sort | Muhamamd Adnan Syed |
| collection | DOAJ |
| description | In person reidentification distance metric learning suffers a great challenge from impostor persons. Mostly, distance metrics are learned by maximizing the similarity between positive pair against impostors that lie on different transform modals. In addition, these impostors are obtained from Gallery view for query sample only, while the Gallery sample is totally ignored. In real world, a given pair of query and Gallery experience different changes in pose, viewpoint, and lighting. Thus, impostors only from Gallery view can not optimally maximize their similarity. Therefore, to resolve these issues we have proposed an impostor resilient multimodal metric (IRM3). IRM3 is learned for each modal transform in the image space and uses impostors from both Probe and Gallery views to effectively restrict large number of impostors. Learned IRM3 is then evaluated on three benchmark datasets, VIPeR, CUHK01, and CUHK03, and shows significant improvement in performance compared to many previous approaches. |
| format | Article |
| id | doaj-art-c1d9086c423b44d9b03ab758a3bd702a |
| institution | Kabale University |
| issn | 1687-5680 1687-5699 |
| language | English |
| publishDate | 2018-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advances in Multimedia |
| spelling | doaj-art-c1d9086c423b44d9b03ab758a3bd702a2025-02-03T01:21:46ZengWileyAdvances in Multimedia1687-56801687-56992018-01-01201810.1155/2018/32024953202495Impostor Resilient Multimodal Metric Learning for Person ReidentificationMuhamamd Adnan Syed0Zhenjun Han1Zhaoju Li2Jianbin Jiao3University of Chinese Academy of Sciences, Beijing, ChinaUniversity of Chinese Academy of Sciences, Beijing, ChinaUniversity of Chinese Academy of Sciences, Beijing, ChinaUniversity of Chinese Academy of Sciences, Beijing, ChinaIn person reidentification distance metric learning suffers a great challenge from impostor persons. Mostly, distance metrics are learned by maximizing the similarity between positive pair against impostors that lie on different transform modals. In addition, these impostors are obtained from Gallery view for query sample only, while the Gallery sample is totally ignored. In real world, a given pair of query and Gallery experience different changes in pose, viewpoint, and lighting. Thus, impostors only from Gallery view can not optimally maximize their similarity. Therefore, to resolve these issues we have proposed an impostor resilient multimodal metric (IRM3). IRM3 is learned for each modal transform in the image space and uses impostors from both Probe and Gallery views to effectively restrict large number of impostors. Learned IRM3 is then evaluated on three benchmark datasets, VIPeR, CUHK01, and CUHK03, and shows significant improvement in performance compared to many previous approaches.http://dx.doi.org/10.1155/2018/3202495 |
| spellingShingle | Muhamamd Adnan Syed Zhenjun Han Zhaoju Li Jianbin Jiao Impostor Resilient Multimodal Metric Learning for Person Reidentification Advances in Multimedia |
| title | Impostor Resilient Multimodal Metric Learning for Person Reidentification |
| title_full | Impostor Resilient Multimodal Metric Learning for Person Reidentification |
| title_fullStr | Impostor Resilient Multimodal Metric Learning for Person Reidentification |
| title_full_unstemmed | Impostor Resilient Multimodal Metric Learning for Person Reidentification |
| title_short | Impostor Resilient Multimodal Metric Learning for Person Reidentification |
| title_sort | impostor resilient multimodal metric learning for person reidentification |
| url | http://dx.doi.org/10.1155/2018/3202495 |
| work_keys_str_mv | AT muhamamdadnansyed impostorresilientmultimodalmetriclearningforpersonreidentification AT zhenjunhan impostorresilientmultimodalmetriclearningforpersonreidentification AT zhaojuli impostorresilientmultimodalmetriclearningforpersonreidentification AT jianbinjiao impostorresilientmultimodalmetriclearningforpersonreidentification |