Understanding visual scenes in 200 msec: Results from Human and Modeling Experiments

August 8, 2006
Aude Oliva | Brain and Cognitive Sciences, MIT

One of the remarkable aspects of human image understanding is that we are able to recognize the meaning of a novel image very quickly and independently of the complexity of the image. This talk will review findings in human perception that help us understand which mechanisms the human brain uses to achieve fast visual recognition, accurate visual search and adequate memorization of visual information. It also will describe the limits of human perception, as well as how to use our understanding of the pros and cons of these mechanisms for designing artificial vision systems and visual displays for human use.

Speaker Details

Since July 2004, I am Assistant Professor in the Department of Brain and Cognitive Sciences at MIT, leading the Computational Visual Cognition Laboratory, a reseach team part of the Perceptual Science Group. After two Masters and a PhD in Cognitive Science, earned at the Institut National Polytechnique de Grenoble (Laboratoire des Images et Signaux), I followed a journey of research and professor assistanship positions in various domains of visual sciences and at various places in the world (Glasgow, Kyoto, Grenoble, Harvard Med. School in Boston, Psychology and Computer Science, Michigan State Uni). I am currently developing three research programs at MIT investigating the psychological, formal and neural substrates of (1) Scene Understanding (2) Visual Complexity and (3) more generally, investigating Computational Visual Cognition. When I am not working on sciences, I create Hybrid Images (for exhibits and for fun), watch science fiction movies or dream I am an astronaute (I was born on the 21 st of July … ring a moon?).