Multi-Proxy Constraint Loss for Vehicle Re-Identification
Vehicle re-identification plays an important role in cross-camera tracking and vehicle search in surveillance videos. Large variance in the appearance of the same vehicle captured by different cameras and high similarity of different vehicles with the same model poses challenges for vehicle re-ident...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-09-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/18/5142 |
| _version_ | 1797554125067517952 |
|---|---|
| author | Xu Chen Haigang Sui Jian Fang Mingting Zhou Chen Wu |
| author_facet | Xu Chen Haigang Sui Jian Fang Mingting Zhou Chen Wu |
| author_sort | Xu Chen |
| collection | DOAJ |
| description | Vehicle re-identification plays an important role in cross-camera tracking and vehicle search in surveillance videos. Large variance in the appearance of the same vehicle captured by different cameras and high similarity of different vehicles with the same model poses challenges for vehicle re-identification. Most existing methods use a center proxy to represent a vehicle identity; however, the intra-class variance leads to great difficulty in fitting images of the same identity to one center feature and the images with high similarity belonging to different identities cannot be separated effectively. In this paper, we propose a sampling strategy considering different viewpoints and a multi-proxy constraint loss function which represents a class with multiple proxies to perform different constraints on images of the same vehicle from different viewpoints. Our proposed sampling strategy contributes to better mine samples corresponding to different proxies in a mini-batch using the camera information. The multi-proxy constraint loss function pulls the image towards the furthest proxy of the same class and pushes the image from the nearest proxy of different class further away, resulting in a larger margin between decision boundaries. Extensive experiments on two large-scale vehicle datasets (VeRi and VehicleID) demonstrate that our learned global features using a single-branch network outperforms previous works with more complicated network and those that further re-rank with spatio-temporal information. In addition, our method is easy to plug into other classification methods to improve the performance. |
| first_indexed | 2024-03-10T16:27:18Z |
| format | Article |
| id | doaj.art-4c73120eddf94f65a4898f9af9a30e63 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T16:27:18Z |
| publishDate | 2020-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-4c73120eddf94f65a4898f9af9a30e632023-11-20T13:07:07ZengMDPI AGSensors1424-82202020-09-012018514210.3390/s20185142Multi-Proxy Constraint Loss for Vehicle Re-IdentificationXu Chen0Haigang Sui1Jian Fang2Mingting Zhou3Chen Wu4The State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaThe State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaCollege of Urban and Environmental Sciences, Central China Normal University, Wuhan 430079, ChinaThe State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan 430079, ChinaNational Engineering Research Center For E-Learning, Central China Normal University, Wuhan 430079, ChinaVehicle re-identification plays an important role in cross-camera tracking and vehicle search in surveillance videos. Large variance in the appearance of the same vehicle captured by different cameras and high similarity of different vehicles with the same model poses challenges for vehicle re-identification. Most existing methods use a center proxy to represent a vehicle identity; however, the intra-class variance leads to great difficulty in fitting images of the same identity to one center feature and the images with high similarity belonging to different identities cannot be separated effectively. In this paper, we propose a sampling strategy considering different viewpoints and a multi-proxy constraint loss function which represents a class with multiple proxies to perform different constraints on images of the same vehicle from different viewpoints. Our proposed sampling strategy contributes to better mine samples corresponding to different proxies in a mini-batch using the camera information. The multi-proxy constraint loss function pulls the image towards the furthest proxy of the same class and pushes the image from the nearest proxy of different class further away, resulting in a larger margin between decision boundaries. Extensive experiments on two large-scale vehicle datasets (VeRi and VehicleID) demonstrate that our learned global features using a single-branch network outperforms previous works with more complicated network and those that further re-rank with spatio-temporal information. In addition, our method is easy to plug into other classification methods to improve the performance.https://www.mdpi.com/1424-8220/20/18/5142sampling considering viewpointsmulti-proxy constraint lossvehicle re-identification |
| spellingShingle | Xu Chen Haigang Sui Jian Fang Mingting Zhou Chen Wu Multi-Proxy Constraint Loss for Vehicle Re-Identification Sensors sampling considering viewpoints multi-proxy constraint loss vehicle re-identification |
| title | Multi-Proxy Constraint Loss for Vehicle Re-Identification |
| title_full | Multi-Proxy Constraint Loss for Vehicle Re-Identification |
| title_fullStr | Multi-Proxy Constraint Loss for Vehicle Re-Identification |
| title_full_unstemmed | Multi-Proxy Constraint Loss for Vehicle Re-Identification |
| title_short | Multi-Proxy Constraint Loss for Vehicle Re-Identification |
| title_sort | multi proxy constraint loss for vehicle re identification |
| topic | sampling considering viewpoints multi-proxy constraint loss vehicle re-identification |
| url | https://www.mdpi.com/1424-8220/20/18/5142 |
| work_keys_str_mv | AT xuchen multiproxyconstraintlossforvehiclereidentification AT haigangsui multiproxyconstraintlossforvehiclereidentification AT jianfang multiproxyconstraintlossforvehiclereidentification AT mingtingzhou multiproxyconstraintlossforvehiclereidentification AT chenwu multiproxyconstraintlossforvehiclereidentification |