Website: http://www.biomotionlab.de

Coming to Canada from

Department of Psychology, Ruhr-Universität-Bochum, Germany

Research involves

Investigating the computational, perceptual and neurophysiological aspects of animate motion perception.

Research relevance

Methods for analysis and synthesis of animate motion will prove useful to motion modelling in computer animation and telecommunications and to clinical movement analysis.

How Our Brains Perceive Animate Motion

When you notice someone in a crowd, your brain performs a range of complex computations, integrating information from a variety of different sources such as the person's facial features, posture, voice, scent-and the way that person moves. The way a person moves reveals a multitude of psychologically and socially relevant attributes. We can recognize friends by the way they walk and we can attribute gender, age, emotional expression, and personality traits to perfect strangers based entirely on their actions, gestures, and facial movements (that is, their animate motion). This ability to recognize different patterns of motion is often overlooked in the study of human perception and the way our visual systems achieve this so effortlessly is still a riddle for vision scientists.

Dr.Nikolaus Troje studies how information is encoded in biological motion patterns and how it can be retrieved from them. His aim is to understand the mechanisms used by the human brain to recognize visual motion. As Canada Research Chair in Vision and Behavioural Science, Dr. Troje combines computational modelling with visual psychophysics and physiological testing techniques. In his experiments, human observers are presented with real, modified or artificially generated displays of human motion patterns and they have to solve tasks such as sex classification, person identification and recognition of emotion. Comparing the performance of human observers with computational models can reveal principles how the brain processes information. Visualizing brain activity with modern imaging techniques offers insight into the involvement of different areas and circuits in the brain.

In-depth knowledge about the way in which humans perceive motion, combined with techniques for generating artificial motion patterns and modifying relevant features of the motion patterns, is the basis for applications in telecommunication and computer animation. To this end, Dr. Troje hopes to develop products that help computer animators generate realistic, psychologically convincing biological motion for computer-generated characters.