Natural Sciences and Engineering Research Council of Canada
Symbol of the Government of Canada

Common menu bar links

Past Winner
1996 E.W.R. Steacie Memorial Fellowship

Demetri Terzopoulos

Computer Science

University of Toronto

When Demetri Terzopoulos asks if you'd like to see his fish, don't be surprised when he turns on his computer. His fish are virtual creatures, and they inhabit a computer-generated virtual world.

Dr. Terzopoulos, a University of Toronto computer science professor, has gained prominence for his outstanding contributions to computer vision and computer graphics and is now also doing pioneering work in artificial life, an emerging field that transcends the traditional boundaries of computer science and biological science.

His virtual fish are remarkable computational models that capture the physics of the animal in its environment, as well as its locomotion, perception, behavior and learning. In the context of computer animation, Dr. Terzopoulos's fish are not just highly realistic graphical puppets - like, for example, the dinosaurs of the hit movie Jurassic Park whose moves were plotted out by skilled human animators. Rather, the fish are autonomous artificial creatures with "eyes" to see their virtual world and "brains" that govern their actions. They swim, forage, eat and mate on their own.

Dr. Terzopoulos sees exciting research opportunities with potentially big pay-offs at the intersection of different disciplines. "Computer vision and computer graphics have developed independently of one another into major fields of computer science," he explains. "However, they are fundamentally related. Graphics involves creating images from models, while vision involves creating models from images. The goal of my visual modeling research is to bridge the gap between these two fields."

Equally exciting is Dr. Terzopoulos's work on human facial modelling. He has produced what is widely recognized as the most realistic biomechanical/expressive model of the human face to date. Modelling faces can play a role in planning reconstructive facial surgery and predicting its results. It is also a central concern for automated face recognition and video compression for teleconferencing.

"There are many interesting questions to explore," Dr. Terzopoulos says. "My research is so multidisciplinary, there's no chance of becoming pigeon-holed. In fact, it sometimes feels as if I'm at the centre of a vortex."