Dr Sabesan Sivapalan

Biography

Dr Sabesan Sivapalan is an Advance Queensland Research fellow in Vision & Signal Processing and the Speech, Audio, Image and Video Technologies group within the Science and Engineering Faculty at QUT.  As a part of the Advance Queensland research fellowship, he is currently leading the research team at  TrademarkVision (A Compu mark company).  Dr Sivapalan received his PhD degree in computer vision from QUT, in December 2014 in a topic of “Human Identification Using Advanced Gait Recognition Approaches”. It is also a part of intelligence surveillance systems in "Airport of the Future Project".
The proposed solutions during this research work contribute to improve the gait-based person recognition performances in various practical scenarios that further enable the adoption of gait recognition into real world security and forensic applications that requires identifying non-intrusive human at distance. The pioneering works on frontal gait recognition, using depth images from Microsoft Kinect, allow gait recognition to be integrated with the biometric portal-based smart gates. Different challenging conditions that affect the human gait have been considered, and the state-of-the art recognition performances have been retrieved.
After finishing his PhD, Dr Sivapalan has been working as a Research Engineer at Queensland Centre For Advanced Technologies (QCAT) within CSIRO. His research work at QCAT mainly focuses on exploring machine learning and computer vision techniques in sports using depth sensors and HeatWave (https://wiki.csiro.au/display/ASL/HeatWave).  In 2016, He joined TrademarkVision as a primary researcher to develop deep learning solutions to trademark search. In 2017, he has secured the Advance Queensland fellowship grant that facilitates the collaboration work with TrademarkVision and QUT.
At Trademark vision, Dr Sivapalan leads research team to provide machine learning and computer vision solutions to do effective content-based image retrieval for trademark search. This involves exploring  and applying cutting-edge deep learning approaches on object detection, text mining, semantic understanding of trademark images, image alignments,  image quality analysis and 3d analysis of industrial design images.