Viewpoint-Specific Representations in Three-Dimensional Object Recognition.

摘要

We report a series of psychological experiments that explore different aspects of the problem of object representation and recognition in human vision. Contrary to the paradigmatic view which holds that the representations are three dimensional and object centered, the results consistently support the notion of view specific representations that include at most partial depth information. In simulated experiments that involved the same stimuli shown to the human subjects, computational models built around two dimensional multiple-view representations replicated our main psychological results, including patterns of generalization errors and the time course of perceptual learning.