Spherical Region-Based Matching of Vanishing Points in Catadioptric Images - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2008

Spherical Region-Based Matching of Vanishing Points in Catadioptric Images


A large literature exists for rotation estimation using van- ishing points (VP). All these VP-based methods require to match the vanishing points in consecutive frames. It is usually considered that this matching step is trivial. However whereas some techniques exist, they suffer from some important limitations and might not work correctly in real robotic applications. For example, the continuity constraint (which aims to match the pair of VPs having the lowest angular distance) cannot handle large rotations and the line matching technique is very slow to execute and assumes accurate/stable line detection. In this paper, we present a fast and robust method to build the correspondences of VPs in catadioptric images. This work is strongly motivated by our research on real-time rotation estimation for dynamic vehicles using catadioptric vision. The underlying idea of the proposed method consists in matching, by histogram comparison, the spherical regions defined by the VPs in the equivalent sphere. Experiments have demonstrated the efficiency of this approach in terms of speed and robustness to translation, rotation, dynamic environment and image blurring.
Fichier principal
Vignette du fichier
A9CR.pdf (1.21 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

inria-00325320 , version 1 (28-09-2008)


  • HAL Id : inria-00325320 , version 1


Jean-Charles Bazin, Inso Kweon, Cedric Demonceaux, Pascal Vasseur. Spherical Region-Based Matching of Vanishing Points in Catadioptric Images. The 8th Workshop on Omnidirectional Vision, Camera Networks and Non-classical Cameras - OMNIVIS, Rahul Swaminathan and Vincenzo Caglioti and Antonis Argyros, Oct 2008, Marseille, France. ⟨inria-00325320⟩
85 View
265 Download


Gmail Facebook X LinkedIn More