Visual Surveillance Algorithm for Dense Crowd
Ge Yang, Zihao Chen · 2019
In this paper, the traditional pedestrian detection and tracking method fails due to the pedestrian occlusion in the dense crowd background. The development of Dense Crowd Visual Tracking algorithm based on Deep Learning (DTDL) is introduced. DTDL is mainly composed of pedestrian detection algorithm and Visual tracking framework. Experimental results show that DTDL is more robust than other tracking algorithms, and its running speed is close to real-time tracking.