20161018-35 李鸿升：Crowd Counting and Behaviour Modeling with...

报告嘉宾1：李鸿升（香港中文大学）

报告时间：2016年10月18日（星期二）晚20:00（北京时间）

报告题目：Crowd Counting and Behaviour Modeling with Convolutional Neural Networks

主持人：  欧阳万里（香港中文大学）

报告摘要：

Counting crowd and modeling crowd behaviours in surveillance scenes with large population density has escalating demands in intelligent surveillance industry. It has drawn increasing attention from the computer vision community in recent years. In this talk, I will first introduce our works on crowd counting that are able to count the number of crowd within a region-of-interst (ROI) or across a line-of-interest (LOI) in crowded scenes. Convolutional Neural Networks with specifically designed network structures and supervision signals are proposed to solve these two challenging tasks. In the second part of my talk, I will introduce our latest work on modeling pedestrian behaviours in crowded scene. A pedestrian behavior encoding scheme is designed to provide a general representation of walking paths, which can be used as the input and output of CNN. The proposed Behavior-CNN is trained with real-scene crowd data and then thoroughly investigated from different aspects. Multiple applications can be built based on the proposed behaviour modeling algorithm.

参考文献：

[1] S. Yi, H. Li, X. Wang, Pedestrian Behavior Understanding and Prediction with Deep Neural Networks, ECCV 2016

[2] Z. Zhao, H. Li, R. Zhao, X. Wang, Crossing-line Crowd Counting with Two-phase Deep Neural Networks, ECCV 2016

[3] C. Zhang, H. Li, X. Wang, X. Yang, Cross-Scene Crowd Counting via Deep Convolutional Neural Networks, CVPR 2015

报告人简介：

Hongsheng Li received the bachelor’s degree in automation from the East China University of Science and Technology in 2006, and the master’s and Ph.D. degrees in computer science from Lehigh University, Pennsylvania, United States in 2010 and 2012, respectively. From 2013 to 2014, he was an associate professor at University of Electronic Science and Technology of China. He is currently a Research Assistant Professor in the Department of Electronic Engineering, The Chinese University of Hong Kong. His research interests include computer vision, medical image analysis, and machine learning.

特别鸣谢本次Webinar主要组织者：

VOOC责任委员：何晖光（中国科学院自动化研究所）

VODB协调理事：张利军（南京大学），章国锋（浙江大学）

[Slides]