May 29, 2026
Ice Hockey Video Analytics Using Deep Learning
Title: Ice Hockey Video Analytics Using Deep Learning
Presenter: David A. Clausi, University of Waterloo
Date: June 3rd, 2026 at 10:30am to 11:30am
Location: ENC 201
Ice hockey presents a uniquely challenging environment for computer vision and video analytics due to its high player density, rapid motion, frequent occlusions, complex interactions, and continuous gameplay. In this talk, I will present a research program focused on advancing automated understanding of ice hockey through modern video analytics techniques. The presentation will cover methods for homography, multi-object detection and tracking, player and puck localization, action recognition from broadcast video. Emphasis will be placed on the challenges associated with unconstrained sports video, including camera motion, scale variation, severe occlusion, and limited annotated training data. This work demonstrates how advances in AI and video understanding are enabling new forms of automated sports intelligence and opening opportunities for impactful interdisciplinary research collaborations.
--
David A. Clausi (PEng, FCAE, FEIC, FIEEE) is a Professor in Systems Design Engineering and University Research Chair at the University of Waterloo who specializes in the field of Intelligent Systems. After earning his Ph.D. (1996) he worked in medical imaging at Mitra Imaging (Waterloo). He started his academic career in 1997 as an Assistant Professor in Geomatics Engineering at the University of Calgary. Dr. Clausi was the Associate Dean - Research & External Partnerships in the Faculty of Engineering (2018-2024). Prof. Clausi has many contributions, conducting research primarily in remote sensing, computer vision, image processing, and sports analytics. He has published extensively, has been an Associate Editor for leading journals, received many scholarships, paper awards, research and teaching excellence awards and his efforts have led to successful commercial implementations.