Attention-Aware Network and Multi-Loss Joint Training Method for Vehicle Re-Identification

Vehicle ReID is a core technology in Intelligent Traffic System (ITS) and an important branch of ReID problem. However, due to the complexity of the traffic environment, the dataset commonly has serious imbalances between positive and negative sample data. To conquer these issues, a novel effective CNN is proposed in this paper. Firstly, we present an Attention-Aware Network that combines the Spatial Attention Model with the Channel Attention Model to make the network focus on high importance areas and further improve the ability to identify matching vehicles. Besides, we propose the Multi-Loss Joint Training strategy to handling the data imbalances. Then to prove the effectiveness of our proposed method. We evaluated our network on the most popular VeRi-776 dataset. Abundant experiment results have shown the effectiveness of our proposed method in vehicle re-identification compared with existing method.

[1]  Dong Liu,et al.  Improving triplet-wise training of convolutional neural network for vehicle re-identification , 2017, 2017 IEEE International Conference on Multimedia and Expo (ICME).

[2]  Liang Zheng,et al.  Re-ranking Person Re-identification with k-Reciprocal Encoding , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Wei Jiang,et al.  A Strong Baseline and Batch Normalization Neck for Deep Person Re-Identification , 2019, IEEE Transactions on Multimedia.

[4]  Koray Kavukcuoglu,et al.  Visual Attention , 2020, Computational Models for Cognitive Vision.

[5]  Xiao Zhang,et al.  Range Loss for Deep Face Recognition with Long-Tailed Training Data , 2016, 2017 IEEE International Conference on Computer Vision (ICCV).

[6]  Yue Xu,et al.  Multi-Attribute Driven Vehicle Re-Identification with Spatial-Temporal Re-Ranking , 2018, 2018 25th IEEE International Conference on Image Processing (ICIP).

[7]  Ching-Tang Hsieh,et al.  Multiple license plate detection for complex background , 2005, 19th International Conference on Advanced Information Networking and Applications (AINA'05) Volume 1 (AINA papers).

[8]  Wenjun Zeng,et al.  Densely Semantically Aligned Person Re-Identification , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  In-So Kweon,et al.  CBAM: Convolutional Block Attention Module , 2018, ECCV.

[10]  Ling-Yu Duan,et al.  Group-Sensitive Triplet Embedding for Vehicle Reidentification , 2018, IEEE Transactions on Multimedia.

[11]  Ben J. A. Kröse,et al.  Keeping Track of Humans: Have I Seen This Person Before? , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.

[12]  Ying Wei,et al.  Combination of Appearance and License Plate Features for Vehicle Re-Identification , 2019, 2019 IEEE International Conference on Image Processing (ICIP).

[13]  Tiejun Huang,et al.  Deep Relative Distance Learning: Tell the Difference between Similar Vehicles , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Ross B. Girshick,et al.  Focal Loss for Dense Object Detection , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Wu Liu,et al.  Large-scale vehicle re-identification in urban surveillance videos , 2016, 2016 IEEE International Conference on Multimedia and Expo (ICME).

[16]  Lucas Beyer,et al.  In Defense of the Triplet Loss for Person Re-Identification , 2017, ArXiv.

[17]  Farzin Aghdasi,et al.  Vehicle Re-identification: an Efficient Baseline Using Triplet Embedding , 2019, 2019 International Joint Conference on Neural Networks (IJCNN).