Improved orientation estimation for texture planes using multiple vanishing points

Citation
E. Ribeiro et Er. Hancock, Improved orientation estimation for texture planes using multiple vanishing points, PATT RECOG, 33(10), 2000, pp. 1599-1610
Citations number
22
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
PATTERN RECOGNITION
ISSN journal
00313203 → ACNP
Volume
33
Issue
10
Year of publication
2000
Pages
1599 - 1610
Database
ISI
SICI code
0031-3203(200010)33:10<1599:IOEFTP>2.0.ZU;2-Q
Abstract
Vanishing point locations play a critical role in determining the perspecti ve orientation of textured planes. However, if only a single: vanishing poi nt is available then the problem is undetermined. The tilt direction must b e computed using supplementary information such as the texture gradient. In this paper we describe a method for multiple vanishing point and, hence co mplete perspective pose estimation, which obviates the need to compute the texture gradient. The method is based on local spectral analysis. It exploi ts the fact that spectral orientation is uniform along lines that radiate f rom vanishing points on the image plane. We experiment with the new method on both synthetic and real-world imagery. This demonstrates that the method provides accurate pose angle estimates, even when the slant angle is large . (C) 2000 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.