报告主题：《A Skeleton Based Action Recognition》
In AI, action recognition is a key application. It helps in retrieving images or video based on actions like sports, fighting or hand signals. In the autonomous driving environment, it becomes very important in recognizing the hand signals from traffic police. In this research, we propose trust gates in the spatio-temporal LSTM network based on skeleton input to improve the action recognition performance. We also propose global context-aware attention network to further improve the performance. Scale selection and feature boosting networks are also introduced to show the effectiveness in action recognition.
Prof. Alex Kot has been with the Nanyang Technological University, Singapore since 1991. He headed the Division of Information Engineering at the School of Electrical and Electronic Engineering for eleven years and served as Associate Chair/ Research and Vice Dean Research for the School of Electrical and Electronic Engineering and eight years as Associate Dean for College of Engineering. He is currently Professor and Director of Rapid-Rich Object SEarch (ROSE) Lab and NTU-PKU Joint Research Institute. He has published extensively in the areas of signal processing for communication, biometrics, image forensics, information security and computer vision and machine learning.
Dr. Kot served as Associate Editor for more than ten journals, mostly for IEEE transactions. He has served the IEEE SP Society in various capacities such as the General Co-Chair for the 2004 IEEE International Conference on Image Processing and the Vice-President for the IEEE Signal Processing Society. He received the Best Teacher of the Year Award and is a co-author for several Best Paper Awards including ICPR, IEEE WIFS and IWDW. He was elected as the IEEE Distinguished Lecturer for the Signal Processing Society and the Circuits and Systems Society. He is a Fellow of IES, a Fellow of IEEE, and a Fellow of Academy of Engineering, Singapore.