Yannick Benezeth

Yannick Benezeth is a Professor of Computer Science at Université de Bourgogne Franche-Comté in Dijon, France, where he teaches courses on databases, optimization, and image/video processing at the IUT de Dijon. He conducts research at the ImViA research laboratory (EA7535), focusing on video health monitoring and video analytics applications. His academic journey includes serving as an Associate Professor from 2011-2024 and earning his Habilitation à Diriger des Recherches (HDR) in 2019.

Dr. Benezeth's research interests center on video-based health monitoring systems, particularly remote photoplethysmography (rPPG) for non-contact vital sign measurement. His work spans computer vision, video analytics, and physiological signal processing, with applications in stress detection, abnormal event recognition, and health monitoring. He has developed several publicly available datasets including UBFC-Phys, UBFC-RPPG, and IMVIA-NIR that have become valuable resources for researchers in affective computing and remote physiological monitoring.

His publication record demonstrates consistent contributions to top computer vision venues including CVPR, ICPR, and IEEE Transactions. Recent work shows a clear trend toward multimodal approaches combining video analysis with physiological signal processing, particularly in psychophysiological stress studies. The UBFC-Phys dataset published in 2021 represents a significant contribution to affective computing research with over 50 participants and comprehensive physiological measurements.

As a research supervisor with HDR qualification, Dr. Benezeth leads projects in the ImViA laboratory focusing on video analytics for healthcare applications. His team has developed innovative methods for background subtraction, abnormal event detection, and skin tissue segmentation that have been adopted by other researchers through his publicly shared code and datasets. Current work appears focused on improving the robustness of video-based physiological measurement under realistic conditions.