Person Re-Identification via Pyramid Multipart Features and Multi-Attention Framework
Video-based person re-identification has become quite attractive due to its importance in many vision surveillance problems. It is a challenging topic due to the inter/intra changes, occlusion, and pose variations involved. In this paper, we propose a pyramid-attentive framework that relies on multi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Big Data and Cognitive Computing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-2289/6/1/20 |
| _version_ | 1797472832728334336 |
|---|---|
| author | Randa Mohamed Bayoumi Elsayed E. Hemayed Mohammad Ehab Ragab Magda B. Fayek |
| author_facet | Randa Mohamed Bayoumi Elsayed E. Hemayed Mohammad Ehab Ragab Magda B. Fayek |
| author_sort | Randa Mohamed Bayoumi |
| collection | DOAJ |
| description | Video-based person re-identification has become quite attractive due to its importance in many vision surveillance problems. It is a challenging topic due to the inter/intra changes, occlusion, and pose variations involved. In this paper, we propose a pyramid-attentive framework that relies on multi-part features and multiple attention to aggregate features of multi-levels and learns attention-based representations of persons through various aspects. Self-attention is used to strengthen the most discriminative features in the spatial and channel domains and hence capture robust global information. We propose the use of part-relation attention between different multi-granularities of features’ representation to focus on learning appropriate local features. Temporal attention is used to aggregate temporal features. We integrate the most robust features in the global and multi-level views to build an effective convolution neural network (CNN) model. The proposed model outperforms the previous state-of-the art models on three datasets. Notably, using the proposed model enables the achievement of 98.9% (a relative improvement of 2.7% on the GRL) top1 accuracy and 99.3% mAP on the PRID2011, and 92.8% (a relative improvement of 2.4% relative to GRL) top1 accuracy on iLIDS-vid. We also explore the generalization ability of our model on a cross dataset. |
| first_indexed | 2024-03-09T20:06:44Z |
| format | Article |
| id | doaj.art-649524bc3e5b4398808aa2a3aee604b6 |
| institution | Directory Open Access Journal |
| issn | 2504-2289 |
| language | English |
| last_indexed | 2024-03-09T20:06:44Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Big Data and Cognitive Computing |
| spelling | doaj.art-649524bc3e5b4398808aa2a3aee604b62023-11-24T00:29:04ZengMDPI AGBig Data and Cognitive Computing2504-22892022-02-01612010.3390/bdcc6010020Person Re-Identification via Pyramid Multipart Features and Multi-Attention FrameworkRanda Mohamed Bayoumi0Elsayed E. Hemayed1Mohammad Ehab Ragab2Magda B. Fayek3Informatics Research Department, Electronics Research Institute, Giza 12622, EgyptComputer Engineering Department, Faculty of Engineering, Cairo University, Giza 12613, EgyptInformatics Research Department, Electronics Research Institute, Giza 12622, EgyptComputer Engineering Department, Faculty of Engineering, Cairo University, Giza 12613, EgyptVideo-based person re-identification has become quite attractive due to its importance in many vision surveillance problems. It is a challenging topic due to the inter/intra changes, occlusion, and pose variations involved. In this paper, we propose a pyramid-attentive framework that relies on multi-part features and multiple attention to aggregate features of multi-levels and learns attention-based representations of persons through various aspects. Self-attention is used to strengthen the most discriminative features in the spatial and channel domains and hence capture robust global information. We propose the use of part-relation attention between different multi-granularities of features’ representation to focus on learning appropriate local features. Temporal attention is used to aggregate temporal features. We integrate the most robust features in the global and multi-level views to build an effective convolution neural network (CNN) model. The proposed model outperforms the previous state-of-the art models on three datasets. Notably, using the proposed model enables the achievement of 98.9% (a relative improvement of 2.7% on the GRL) top1 accuracy and 99.3% mAP on the PRID2011, and 92.8% (a relative improvement of 2.4% relative to GRL) top1 accuracy on iLIDS-vid. We also explore the generalization ability of our model on a cross dataset.https://www.mdpi.com/2504-2289/6/1/20computer visiondeep learningperson re-identificationattentiontemporal aggregationmulti-granularities |
| spellingShingle | Randa Mohamed Bayoumi Elsayed E. Hemayed Mohammad Ehab Ragab Magda B. Fayek Person Re-Identification via Pyramid Multipart Features and Multi-Attention Framework Big Data and Cognitive Computing computer vision deep learning person re-identification attention temporal aggregation multi-granularities |
| title | Person Re-Identification via Pyramid Multipart Features and Multi-Attention Framework |
| title_full | Person Re-Identification via Pyramid Multipart Features and Multi-Attention Framework |
| title_fullStr | Person Re-Identification via Pyramid Multipart Features and Multi-Attention Framework |
| title_full_unstemmed | Person Re-Identification via Pyramid Multipart Features and Multi-Attention Framework |
| title_short | Person Re-Identification via Pyramid Multipart Features and Multi-Attention Framework |
| title_sort | person re identification via pyramid multipart features and multi attention framework |
| topic | computer vision deep learning person re-identification attention temporal aggregation multi-granularities |
| url | https://www.mdpi.com/2504-2289/6/1/20 |
| work_keys_str_mv | AT randamohamedbayoumi personreidentificationviapyramidmultipartfeaturesandmultiattentionframework AT elsayedehemayed personreidentificationviapyramidmultipartfeaturesandmultiattentionframework AT mohammadehabragab personreidentificationviapyramidmultipartfeaturesandmultiattentionframework AT magdabfayek personreidentificationviapyramidmultipartfeaturesandmultiattentionframework |