Fitting 3D models to catadioptric images

Contact : Eric Marchand
Created : August 22, 2006
Last Modification : January 4, 2007

Overview

This page presents a real-time, robust and efficient 3D model-based tracking algorithm adapted to catadioptric images. A virtual visual servoing approach is used for 3D tracking. This method is similar to more classical non-linear pose computation techniques. Robustness is obtained by integrating a M-estimator into the virtual visual control law via an iteratively re-weighted least squares implementation.

Results of the virtual visual servoing tracker with perspective cameras are presented in Real-time 3D localisation and tracking page.

In this page we present results of the extension of this approach to the use of catadioptric cameras.

Reference

Video results

click for the video (Thanks to Guillaume Caron)

Tracking a box Tracking a box Tracking 2 plinths
Output results 1 test test
Image processing test test

Acknowledgments

The firts video has been provided by Guillaume Caron, MIS laboratory (Université de Picardie Jules Verne). The latter images sequences and camera calibration parameters have been provided by Hicham Hadj-Abdelkader from Lasmea.

| Lagadic | Map | Team | Publications | Demonstrations |
Irisa - Inria - Copyright 2009 © Lagadic Project