Real-time Quadrifocal Visual Odometry

Andrew Comport 1 Ezio Malis 2 Patrick Rives 3
2 AROBAS - Advanced Robotics and Autonomous Systems
CRISAM - Inria Sophia Antipolis - Méditerranée
3 Lagadic - Visual servoing in robotics, computer vision, and augmented reality
CRISAM - Inria Sophia Antipolis - Méditerranée , Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : This paper describes a new image-based approach to tracking the 6 degrees of freedom trajectory of a stereo camera pair using a corresponding reference image pair whilst simultaneously determining pixel matching between onsecutive images in a sequence. A dense minimisation approach is employed which directly uses all grey-scale information available within the stereo pair (or stereo region) leading to very robust and precise results. Metric 3d structure constraints are imposed by consistently warping corresponding stereo images to generate novel viewpoints at each stereo acquisition. An iterative non-linear trajectory estimation approach is formulated based on a quadrifocal relationship between the image intensities with adjacent views of the stereo pair. A robust M-estimation technique is used to reject outliers corresponding to moving objects within the scene or other outliers such as occlusions and illumination changes. The technique is applied to recovering the trajectory of a moving vehicle in long and difficult sequences of images.
Document type :
Journal articles
Liste complète des métadonnées

Cited literature [33 references]  Display  Hide  Download

https://hal.inria.fr/hal-00766839
Contributor : Patrick Rives <>
Submitted on : Wednesday, December 19, 2012 - 10:27:39 AM
Last modification on : Thursday, February 7, 2019 - 3:52:26 PM
Document(s) archivé(s) le : Wednesday, March 20, 2013 - 11:31:52 AM

File

ijrr08.pdf
Files produced by the author(s)

Identifiers

Citation

Andrew Comport, Ezio Malis, Patrick Rives. Real-time Quadrifocal Visual Odometry. International Journal of Robotics Research, SAGE Publications, 2010, 29 (2), pp.245-266. ⟨10.1177/0278364909356601⟩. ⟨hal-00766839⟩

Share

Metrics

Record views

875

Files downloads

1039