Abstract | ||
---|---|---|
Person re-identification is a hot topic due to its huge application potentials. Siamese network is a good method to learn feature representation in verification tasks and has been used in previous person re-identification research, but hard to convergence during training process. This paper presents a multi-task learning pipeline including Siamese loss for learning deep feature representations of people appearance. Firstly, we point out the defects of training a convolutional neural network (CNN) only with Siamese loss which is usually used for person re-identification. Secondly, a multi-task CNN for person re-identification combing the Softmax loss with Siameses loss is proposed. Finally, some experiments are carried out to test the performance of proposed multi-task person appearance learning pipeline. Experiments on various pedestrian dataset shows the effectiveness of our pipeline. Our method outperforms state-of-the-art person re-identification methods in some public datasets. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1007/978-3-319-67777-4_23 | INTELLIGENCE SCIENCE AND BIG DATA ENGINEERING, ISCIDE 2017 |
Keywords | Field | DocType |
Person re-identification, Siamese network, Multi-task learning, Convolutional neural network | Convergence (routing),Pedestrian,Multi-task learning,Softmax function,Convolutional neural network,Computer science,Artificial intelligence,Combing,Machine learning | Conference |
Volume | ISSN | Citations |
10559 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 19 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hua Gao | 1 | 0 | 1.35 |
Lingyan Yu | 2 | 0 | 0.34 |
Yujiao Huang | 3 | 1 | 1.03 |
Yiwei Dong | 4 | 0 | 0.34 |
Sixian Chan | 5 | 12 | 7.69 |