Visual Similarity, Judgmental Certainty and Stereo Correspondence

Normal human vision is nearly infallible in modeling the visually sensed physical environment in which it evolved. In contrast, most currently available computer vision systems fall far short of human performance in this task, and further, they are generally not capable of being able to assert the correctness of their judgments. In computerized stereo matching systems, correctness of the similarity/identity-matching is almost never guaranteed. In this paper, we explore the question of the extent to which judgments of similarity/identity can be made essentially error-free in support of obtaining a relatively dense depth model of a natural outdoor scene. We argue for the necessity of simultaneously producing a crude scene-speci c semantic \overlay". For our experiments, we designed a wavelet-based stereo matching algorithm and use \classi cation-trees" to create a primitive semantic overlay of the scene. A series of mutually independent lters has been designed and implemented based on the study of di erent error sources. Photometric appearance, camera imaging geometry and scene constraints are utilized in these lters. When tested on di erent sets of stereo images, our system has demonstrated above 98% correctness on asserted matches. Finally, we provide a principled basis for relatively dense depth recovery.