We describe a method to solve stereo correspondence using controlled eye (or camera) movements. Eye movements supply additional image frames and monocular depth estimate, which can be used to constrain stereo matching. Because the eye movements are small, traditional stereo techniques of stereo with multiple frame will not work. We develop an alternative approach using a systematic analysis to define a probability distribution for the errors. Our matching strategy then matches the most probable points first, thereby reducing the ambiguity for the remaining matches. We demonstrate this algorithms with several examples.
ASJC Scopus subject areas
- Computer Science(all)