A. B. Murynin ,   I. A. Matveev ,   V. N. Kolmogorov Graphicon-99 , 09.1999 , language: английский  Аннотация One of the crucial problems in machine vision is reconstruction of 3D shape of objects from their images and the most complicated aspect of this problem is disparity estimation. Some techniques of disparity calculation in realtime monocular and binocular environment are presented. The algorithms have been developed on the basis of correlation method estimating corresponding points on images in stereo set. Performance optimization is achieved by some modifications of correlation method including a pyramidal presentation of the images, by introducing some constraints on the conditions of image recording, and by involving an a priory information about objects in study. The method for estimating the size and shape of correlation region is proposed. The system was tested on a database of stereo-images.  Ключевые слова Computer vision, stereoscopic vision, surface reconstruction, pyramidal data representation.   Полный текст     в формате zip