Gabriel Kreiman, Ph.D.
Associate Professor of Ophthalmology
F.M. Kirby Neurobiology Center
Center for Life Science, Room 12-253
3 Blackfan Circle
Boston, MA 02115
Visit my lab page here.
Our lab is interested in the neuronal circuits and algorithms responsible for visual object recognitionand memory formation. Visual object recognition is crucial for most everyday tasks including face identification, reading and navigation. In addition, many Machine Vision applications such as surveillance, biomedical image interpretation and automatic navigation rely strongly on visual pattern recognition. Through millions of years of evolution, the primate visual system has achieved a highly selective, rapid and robust recognition machinery. Selectivity refers to discriminating among multiple similar objects (e.g. face identification). Robustness refers to recognizing an object in spite of transformations at the pixel level (e.g. changes in viewpoint, scale, position, clutter, contrast, color, illumination, etc.). Remarkably, visual selectivity and robustness can be achieved in only ~150 ms of processing, placing a strong constraint on the number of computations that the cortical circuitry can perform to achieve immediate visual recognition. Our lab combines neurophysiology, psychophysics and theoretical/computational modeling to understand the neuronal circuits, algorithms and computations performed by the visual system and we use this knowledge to develop biophysically-inspired computational approaches to machine vision.
For a complete listing of publications click here.
Last Update: 11/7/2013