Humans recognize visual objects with such ease that it is easy to overlook what an impressive computational feat this represents. Any given object in the world can cast an effectively infinite number of different images onto the retina, depending on its position relative to the viewer, the configuration of light sources, and the presence of other objects in the visual field. In spite of this extreme variation, biological visual systems are able to effortlessly recognize at least hundreds of thousands of distinct object classes—a feat that no current artificial system can come close to achieving.
Our laboratory seeks to understand the computational underpinnings of object recognition through a concerted effort on two fronts. First, we endeavor to understand the workings of biological visual systems using a variety of experimental techniques, ranging from microelectrode recordings to visual psychophysics. Second, we attempt to instantiate what we have learned into artificial object recognition systems, leveraging recent advances in parallel computing to build systems that begin to approach the scale of natural systems. By combining reverse- and forward-engineering approaches, we hope to accelerate progress in both domains.