Skip to main content

November 14, 2012

Speaker: Gregory Dudek, Professor and Director, School of Computer Science, James McGill Chair, McGill University

Title: Automated Video Summarization and the Detection of Notable Events

Abstract: This talk will address the automated analysis of video data, and specifically video data collected underwater with an amphibious vehicle (the Aqua2 hexapod). Automated systems can collect data at prodigious rates and the timely analysis of this data is a growing challenge, especially when there are bandwidth constraints between the data source and the people who must examine the data. We are specifically interested in the real-time summarization and detection of the most interesting events in a video sequence, for use by humans who will analyze the data either in real time or offline. To do this, we are developing methods that adapt to video data streams in real time to collect salient events and using them in the context of a group of vehicles that fly, swim and float.

University of Ontario Institute of Technology logo