Abstract | ||
---|---|---|
Person re-identification aims at identifying a certain person across non-overlapping multi-camera networks. It is a fundamental and challenging task in automated video surveillance. Most existing researches mainly rely on hand-crafted features, resulting in unsatisfactory performance. In this paper, we propose a multi-scale triplet convolutional neural network which captures visual appearance of a person at various scales. We propose to optimize the network parameters by a comparative similarity loss on massive sample triplets, addressing the problem of small training set in person re-identification. In particular, we design a unified multi-scale network architecture consisting of both deep and shallow neural networks, towards learning robust and effective features for person re-identification under complex conditions. Extensive evaluation on the real-world Market-1501 dataset have demonstrated the effectiveness of the proposed approach.
|
Year | DOI | Venue |
---|---|---|
2016 | 10.1145/2964284.2967209 | MM '16: ACM Multimedia Conference
Amsterdam
The Netherlands
October, 2016 |
Field | DocType | ISBN |
Training set,Computer vision,Computer science,Convolutional neural network,Network architecture,Artificial intelligence,Artificial neural network,Machine learning,Visual appearance | Conference | 978-1-4503-3603-1 |
Citations | PageRank | References |
44 | 1.15 | 16 |
Authors | ||
7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jiawei Liu | 1 | 88 | 8.41 |
Zheng-Jun Zha | 2 | 2822 | 152.79 |
Qi Tian | 3 | 6443 | 331.75 |
Dong Liu | 4 | 721 | 74.92 |
Ting Yao | 5 | 842 | 52.62 |
Qiang Ling | 6 | 251 | 37.43 |
Tao Mei | 7 | 4702 | 288.54 |