Vision using two optic sensors (i.e., eyes if you're an animal, cameras if you're a robot). A single view can reveal information about objects through shape, texture, and shading; multiple views add depth.

Stereo vision is accomplished by matching points and contours from one image to another. The disparity between their locations, combined with a knowledge of the geometric relation between the two eyes (or cameras), allows the distance to the viewer to the object to be recovered. This computation can be done fairly easily by a computer program; what's hard is the matching of points and contours between images. (See also computer vision.)

Log in or register to write something here or to contact authors.