Object Recognition by a Cascade of Edge Probes - Robotics Institute Carnegie Mellon University

Object Recognition by a Cascade of Edge Probes

Conference Paper, Proceedings of British Machine Vision Conference (BMVC '02), pp. 103 - 112, September, 2002

Abstract

We frame the problem of object recognition from edge cues in terms of determining whether individual edge pixels belong to the target object or to clutter, based on the configuration of edges in their vicinity. A classifier solves this problem by computing sparse, localized edge features at image locations determined at training time. In order to save computation and solve the aperture problem, we apply a cascade of these classifiers to the image, each of which computes edge features over larger image regions than its predecessors. Experiments apply this approach to the recognition of real objects with holes and wiry components in cluttered scenes under arbitrary out-of-image-plane rotation.

BibTeX

@conference{Carmichael-2002-8536,
author = {Owen Carmichael and Martial Hebert},
title = {Object Recognition by a Cascade of Edge Probes},
booktitle = {Proceedings of British Machine Vision Conference (BMVC '02)},
year = {2002},
month = {September},
editor = {Paul Rosin, David Marshall},
pages = {103 - 112},
publisher = {British Machine Vision Association},
keywords = {object recognition, computer vision},
}