Skip to Main content Skip to Navigation
Conference papers

Calibrage et Reconstruction à l'aide de Parallélépipèdes et de Parallélogrammes

Marta Wilczkowiak 1 Peter Sturm 1 Edmond Boyer 1
1 MOVI - Modeling, localization, recognition and interpretation in computer vision
GRAVIR - IMAG - Graphisme, Vision et Robotique, Inria Grenoble - Rhône-Alpes, CNRS - Centre National de la Recherche Scientifique : FR71
Abstract : In this paper, efficient tools for reconstruction from a single or a few images are presented. They exploit geometric constraints frequently present in man-made environments and allow camera calibration as well as scene structure to be estimated with few interactions and little a priori knowledge. The proposed approach is based on primitives that naturally characterize rigidity constraints: parallelepipeds and parallelograms. The intrinsic metric characteristics of these primitives appear to be dual to the intrinsic characteristics of a perspective camera. Thus any knowledge on either a primitive or a camera can be used to estimate the respective dual entity. We describe the constraints that can be derived in the single image context as well as when several images are available. These principles are illustrated by examples of reconstructions and augmented reality using uncalibrated images.
Document type :
Conference papers
Complete list of metadatas

Cited literature [22 references]  Display  Hide  Download

https://hal.inria.fr/inria-00525647
Contributor : Peter Sturm <>
Submitted on : Thursday, May 26, 2011 - 1:54:24 PM
Last modification on : Friday, June 26, 2020 - 4:04:03 PM
Long-term archiving on: : Saturday, August 27, 2011 - 2:23:16 AM

Files

WilczkowiakSturmBoyer-rfia02.p...
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00525647, version 1

Collections

CNRS | INRIA | IMAG | AFIA | UGA

Citation

Marta Wilczkowiak, Peter Sturm, Edmond Boyer. Calibrage et Reconstruction à l'aide de Parallélépipèdes et de Parallélogrammes. 13ème congrès francophone de Reconnaissance des formes et d'Intelligence artificielle (RFIA '02), Jan 2002, Angers, France. pp.849-857. ⟨inria-00525647⟩

Share

Metrics

Record views

190

Files downloads

308