Professor Greg Mori from the School of Computing Science at Simon Fraser University,recently won the 2008 Award for Research Excellence and Service to the Research Community from the Canadian Image Processing and Pattern Recognition Society (CIPPRS).

Professor Mori’s work is based on computer vision with a particular focus on problems of human activity recognition and human body pose estimation.

His research has wide applications in the fields of surveillance and security. In surveillance, the algorithm that Dr. Mori is building attempts to automatically detect suspicious activity thus eliminating the need to sit through and watch camera recordings. Mori explains, “The particular focus of my work is to figure out where people are in images or videos, and then to figure out what activities they are performing. The combination of activities – doing things that are unusual with respect to everyone else or the environment could be classified as interesting or suspicious activity”. Mori’s research would also enable a human operator to query the video. For example, you could find out the number of times anyone stands in the parking lot for more than 5 minutes. Another application of human activity recognition is human computer interaction. For example, The Sony Eye-toy is a motion sensitive camera that uses video footage of the gamer to control the game. According to Professor Mori, this PlayStation 2 device is relatively primitive with what it can do with regards to interpreting action. He hopes that his research develops real time versions for using such applications. What that means is that “the algorithm would be able to tell the difference between someone waving his left hand versus the right, bowing versus jumping, the different boxing motions etc.”

Another computer graphic application that Professor Mori is working on is Motion Synthesis. Computer games normally use realistic 3D models of human beings. Animations are created by bending and configuring joints and manipulating joint angles in the 3D models. The motion synthesis approach that Professor Mori and his team are working on actually takes real video frames and uses those to generate animations rather than a 3D model.

To find out more about Dr. Mori’s exciting research, please visit the Vision & Media Lab

Written by Salima Vastani