Neural Basis of Motion Perception
Abstract
Our research is concerned with 'high level' vision with a strong biological slant. The last two or three decades have seen breathtaking progress in the three disciplines--cognitive psychology, AI and visual neurophysiology-- but they have been pursued more or less independently. We believe that the time is now ripe for forging links between these disciplines for an integrated approach to vision. We have had two goals in mind: (1) To develop conceptual links between neurophysiology and perception; (2) To develop specific tests for computational models of human vision. Our research has called into question several widely accepted dogmas concerning the mechanisms of early vision. Also; we have been able to discover several novel visual phenomena (e.g., motion capture, stereo-capture, etc.) and have identified a wide range of new 'natural constraints' that govern the perception of shape-from shading structure from motion and motion correspondence. Also, we have discovered striking perceptual correlates of several well-known physiological observations (e.g., 'phantom contours'-stimuli which selectively activate the magnocellular pathway.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 31, 1992
- Accession Number
- ADA261452
Entities
People
- V. S. Ramachandran
Organizations
- University of California, San Diego