A study on location information aided re-identification in CCTV environment

The video data taken by CCTV cameras are becoming abundant as CCTV is widely spreading in surveillance area. As video analytic technologies are developing, object detection and classification of video data can be performed by machines like CCTV cameras and computers. The automatic object detection and classification can be applied to identify same person from multiple video clips taken at different time or by different cameras. The accuracy of video object classification can be enhanced by using other information. In this paper, a method is introduced, which enhances the accuracy of the person re-identification using location information of CCTV and smart phone.

[1]  Su-Wan Park,et al.  EVM: A New Methodology for Evidential Video Management in Digital CCTV Systems , 2012 .

[2]  Jenq-Neng Hwang,et al.  An ensemble of invariant features for person re-identification , 2015, 2015 IEEE 17th International Workshop on Multimedia Signal Processing (MMSP).