6533b838fe1ef96bd12a4e90

RESEARCH PRODUCT

New systems for extracting 3-D shape information from images

subject

description

Neural architectures may offer an adequate way to deal with early vision since they are able to learn shape features or classify unknown shapes, generalising the features of a few meaningful examples, with a low computational cost after the training phase. Two different neural approaches are proposed by the authors: the first one consists of a cascaded architecture made up by a first stage named BWE (Boundary Webs Extractor) which is aimed to extract a brightness gradient map from the image, followed by a backpropagation network that estimates the geometric parameters of the object parts present in the perceived scene. The second approach is based on the extraction of the boundary webs map from the image and its comparison with boundary webs maps exhaustively generated from synthetic superquadrics. A purposely defined error figure has been used to find the best match between the two kinds of maps. A functional comparison between the two systems is described and the quite satisfactory experimental results are presented.