Malik Opens 2013-2014 Dertouzos Lecturer Series with Talk on Computer Vision
11 October 2013
Jitendra Malik, the Arthur J. Chick Professor of Electrical Engineering and Computer Science at the University of California at Berkeley, kicked off the 2013-2014 Dertouzos Lecturer Series on Thursday, October 10. Malik, who is renowned for his work in computer vision, discussed “The Three R's of Computer Vision: Recognition, Reconstruction and Reorganization” and his view on what progress is needed to further advance the field.
Following an introduction by CSAIL Director Daniela Rus, Malik explained that he was originally convinced to work in the field of computer vision by Rodney Brooks - former CSAIL director and the founder, chairman and CTO of Rethink Robotics – who encouraged him to transition from the field of artificial intelligence. Little did Malik realize, he joked, what seminal Brooks would conduct in the field of AI in the years since that conversation.
Malik’s talk covered the history of computer vision and future research directions. Presenting his talk at CSAIL in 2013 was particularly fitting, he explained, as many consider 2013 to mark 50 years of research in computer vision. The field had its inception in the 1960s, according to Malik, with the beginning of artificial intelligence research, image processing and pattern recognition. While there have been considerable advances in computer vision over the last two decades such as work with face detection, reconstructing three-dimensional models of cities and sensing for control of robots and cars, there are many areas in which computers still perform significantly below human perception.
Malik argued that in order to order to improve present computer vision systems researchers need to continue investigation into the three “R’s” of computer vision – recognition, reconstruction and reorganization – and, in particular, the interaction between these processes. Visual systems do not operate as silos he explained, they work together.
“I think we’re going to see a lot of action in the next few years” between these areas, Malik said. For example, in order for robots to operate successfully in human environments and traverse different spaces, researchers will need to combine recognition techniques with reorganization techniques. Malik gave a preview of his work to create a unified framework for the three R’s of computer vision.
For more information on Malik’s work, please visit: http://www.cs.berkeley.edu/~malik/. For more information on the Dertouzos Lecturer Series, please visit: https://calendar.csail.mit.edu/seminar_series/6770.
-Story by Abby Abazorius, CSAIL