Abstract
Traditional supervisory person re-id technology learning methods mainly relies on pre-marked image data, but there are a lot of unlabeled data in actual security scene, which seriously limits the application of person re-id technology in security monitoring field. Therefore, it is very important to study the semi-supervised learning of unlabeled data generated by antagonistic network. In the process of using GAN to generate data, in this paper we use global and local information of pedestrian image to generate realistic pedestrian image conditionally, and trains robust feature representations for different intra-class changes of cameras, so as to improve the accuracy of person re-id. The experimental results show that this method is more effective than the benchmark method. The performance of dataset Market1501 and Duke MTMC-reID improved by 4% and 3% respectively.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.