Biblio

Person re-identification(Person Re-ID) means that images of a pedestrian from cameras in a surveillance camera network can be automatically retrieved based on one of this pedestrian's image from another camera. The appearance change of pedestrians under different cameras poses a huge challenge to person re-identification. Person re-identification systems based on deep learning can effectively extract the appearance features of pedestrians. In this paper, the feature enhancement experiment is conducted, and the result showed that the current person reidentification datasets are relatively small and cannot fully meet the need of deep training. Therefore, this paper studied the method of using generative adversarial network to extend the person re-identification datasets and proposed a label smoothing regularization for outliers with weight (LSROW) algorithm to make full use of the generated data, effectively improved the accuracy of person re-identification.

We investigate how to protect features corresponding to personal information using homomorphic encryption when matching people in several camera views. Homomorphic encryption can compute a distance between features without decryption. Thus, our method is able to use a computing server on a public network while protecting personal information. To apply homomorphic encryption, our method uses linear quantization to represent each element of the feature as integers. Experimental results show that there is no significant difference in the accuracy of person re-identification with or without homomorphic encryption and linear quantization.

Although the deep learning technology effectively improves the effect of person re-identification (re-ID) in video surveillance, there is still a lack of efficient framework in practical, especially in terms of computational cost, which usually requires GPU support. So this paper explores to solve the actual running performance and an effective person re-ID framework is proposed. A tiny network is designed for person detection and a triplet network is adopted for training feature extraction network. The motion detection and person detection is combined to speed up the whole process. The proposed framework is tested in practice and the results show that it can run in real-time on an ordinary PC machine. And the accuracy achieves 91.6% in actual data set. It has a good guidance for person re-ID in actual application.