Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification
Vehicle re-identification is attracting an increasing amount of attention in intelligent transportation and is widely used in public security. In comparison to person re-identification, vehicle re-identification is more challenging because vehicles with different IDs are generated by a unified pipel...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/7/2474 |
| _version_ | 1797571457046282240 |
|---|---|
| author | Honglie Wang Shouqian Sun Lunan Zhou Lilin Guo Xin Min Chao Li |
| author_facet | Honglie Wang Shouqian Sun Lunan Zhou Lilin Guo Xin Min Chao Li |
| author_sort | Honglie Wang |
| collection | DOAJ |
| description | Vehicle re-identification is attracting an increasing amount of attention in intelligent transportation and is widely used in public security. In comparison to person re-identification, vehicle re-identification is more challenging because vehicles with different IDs are generated by a unified pipeline and cannot only be distinguished based on the subtle differences in their features such as lights, ornaments, and decorations. In this paper, we propose a local feature-aware Siamese matching model for vehicle re-identification. A local feature-aware Siamese matching model focuses on the informative parts in an image and these are the parts most likely to differ among vehicles with different IDs. In addition, we utilize Siamese feature matching to better supervise our attention. Furthermore, a perspective transformer network, which can eliminate image deformation, has been designed for feature extraction. We have conducted extensive experiments on three large-scale vehicle re-ID datasets, i.e., VeRi-776, VehicleID, and PKU-VD, and the results show that our method is superior to the state-of-the-art methods. |
| first_indexed | 2024-03-10T20:40:56Z |
| format | Article |
| id | doaj.art-510e6e6b41e949d980f056946c490d17 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T20:40:56Z |
| publishDate | 2020-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-510e6e6b41e949d980f056946c490d172023-11-19T20:40:27ZengMDPI AGApplied Sciences2076-34172020-04-01107247410.3390/app10072474Local Feature-Aware Siamese Matching Model for Vehicle Re-IdentificationHonglie Wang0Shouqian Sun1Lunan Zhou2Lilin Guo3Xin Min4Chao Li5College of Computer Science and Technology, Zhejiang University, No. 38 Zheda Road, Hangzhou 310027, ChinaCollege of Computer Science and Technology, Zhejiang University, No. 38 Zheda Road, Hangzhou 310027, ChinaInstitute of Advanced Digital Technology and Instrument, Zhejiang University, No. 38 Zheda Road, Hangzhou 310027, ChinaAlibaba Group, Ali Yun Feitian Park, Zhuan Tang Street, West Lake District, Hangzhou 310000, ChinaCollege of Computer Science and Technology, Zhejiang University, No. 38 Zheda Road, Hangzhou 310027, ChinaCollege of Computer Science and Technology, Zhejiang University, No. 38 Zheda Road, Hangzhou 310027, ChinaVehicle re-identification is attracting an increasing amount of attention in intelligent transportation and is widely used in public security. In comparison to person re-identification, vehicle re-identification is more challenging because vehicles with different IDs are generated by a unified pipeline and cannot only be distinguished based on the subtle differences in their features such as lights, ornaments, and decorations. In this paper, we propose a local feature-aware Siamese matching model for vehicle re-identification. A local feature-aware Siamese matching model focuses on the informative parts in an image and these are the parts most likely to differ among vehicles with different IDs. In addition, we utilize Siamese feature matching to better supervise our attention. Furthermore, a perspective transformer network, which can eliminate image deformation, has been designed for feature extraction. We have conducted extensive experiments on three large-scale vehicle re-ID datasets, i.e., VeRi-776, VehicleID, and PKU-VD, and the results show that our method is superior to the state-of-the-art methods.https://www.mdpi.com/2076-3417/10/7/2474vehicle re-identificationattention mechanismsiamese neural networks |
| spellingShingle | Honglie Wang Shouqian Sun Lunan Zhou Lilin Guo Xin Min Chao Li Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification Applied Sciences vehicle re-identification attention mechanism siamese neural networks |
| title | Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification |
| title_full | Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification |
| title_fullStr | Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification |
| title_full_unstemmed | Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification |
| title_short | Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification |
| title_sort | local feature aware siamese matching model for vehicle re identification |
| topic | vehicle re-identification attention mechanism siamese neural networks |
| url | https://www.mdpi.com/2076-3417/10/7/2474 |
| work_keys_str_mv | AT hongliewang localfeatureawaresiamesematchingmodelforvehiclereidentification AT shouqiansun localfeatureawaresiamesematchingmodelforvehiclereidentification AT lunanzhou localfeatureawaresiamesematchingmodelforvehiclereidentification AT lilinguo localfeatureawaresiamesematchingmodelforvehiclereidentification AT xinmin localfeatureawaresiamesematchingmodelforvehiclereidentification AT chaoli localfeatureawaresiamesematchingmodelforvehiclereidentification |