University of Alberta Dictionary of Cognitive Science: Nearest Neighbor Principle

One constraint that can be applied to the motion correspondence problem is called the nearest neighbor principle. The visual system prefers to assign correspondence matches that represent short element displacements (Burt & Sperling, 1981; Ullman, 1979). In other words, short motion correspondence matches are to be preferred over longer motion correspondence matches.

The nearest neighbor principle is a natural constraint because it arises from the geometry of the typical viewing conditions for motion (Ullman, 1979, pp. 114-118). When movement in a three-dimensional world is projected onto a two-dimensional surface (e.g., the retina), slower movements occur with much higher probability on the retina than do faster movements. A preference for slower movement is equivalent to exploiting the nearest neighbor principle, because a short correspondence match represents slow motion, while a long correspondence match represents fast motion.

References: