Distinguished Fellow To Talk on Real-Time Human Action Recognition

Date posted

1 June 2017

It is established that there are recognition rate limitations when using a single modality sensor to perform human action or gesture recognition for human-computer interaction applications. The work presented in this talk addresses the simultaneous utilization of two cost-effective sensors that provide 3D data (a Microsoft Kinect depth camera and a wearable inertial sensor) for the purpose of achieving more robust human action or gesture recognition under realistic operating conditions. The use of computationally efficient features as well as computationally efficient classification and fusion approaches are discussed. Experimental results obtained indicate that the developed fusion framework generates higher recognition rates compared to the situations when each sensor is used individually. Moreover, an actual working action and gesture recognition system has been developed which runs in real-time on any modern laptop without using any dedicated hardware. The developed fusion system is also applied to a medical application called Senior Fitness Test.

Nasser Kehtarnavaz is Erik Jonsson Distinguished Professor in the Department of Electrical and Computer Engineering at the University of Texas at Dallas. His research areas include signal and image processing, real-time implementation on embedded processors, biomedical image analysis, and machine learning. He has authored or co-authored 10 books and more than 340 journal papers/conference papers/patents/industry manuals/editorials in these areas. He is a Fellow of IEEE, a Fellow of SPIE, a Professional Engineer, and Editor-in-Chief of Journal of Real-Time Image Processing. More information on Dr. Kehtarnavaz’s research and teaching activities are provided at


The host of this Distinguished Visiting Fellow is Dr Qi Wang, University of the West of Scotland