Stereo Correspondence Through Feature Grouping and Maximal Cliques
IEEE Transactions on Pattern Analysis and Machine Intelligence
Robust recovery of the epipolar geometry for an uncalibrated stereo rig
ECCV '94 Proceedings of the third European conference on Computer vision (vol. 1)
CVGIP: Image Understanding
Computational cross ratio for computer vision
CVGIP: Image Understanding
International Journal of Computer Vision
Probabilistic analysis of the application of the cross ratio to model based vision
International Journal of Computer Vision
Planar object recognition using projective shape representation
International Journal of Computer Vision
Quantitative planar region detection
International Journal of Computer Vision
Iterative pose estimation using coplanar feature points
Computer Vision and Image Understanding
Projectively invariant decomposition and recognition of planar shapes
International Journal of Computer Vision - Special issue: machine vision research at the Royal Institute of Technology
Relaxation labeling networks for the maximum clique problem
Journal of Artificial Neural Networks - Special issue: neural networks for optimization
The Dynamics of Nonlinear Relaxation Labeling Processes
Journal of Mathematical Imaging and Vision
What can be seen in three dimensions with an uncalibrated stereo rig
ECCV '92 Proceedings of the Second European Conference on Computer Vision
Tracking 3D Coplanar Points in the Invariant Perspective Coordinates Plane
ICPR '96 Proceedings of the 1996 International Conference on Pattern Recognition (ICPR '96) Volume I - Volume 7270
Direct Methods for Evaluating the Planarity and Rigidity of a Surface Using Only 2D Views
ICPR '96 Proceedings of the 1996 International Conference on Pattern Recognition (ICPR '96) Volume I - Volume 7270
ICCV '98 Proceedings of the Sixth International Conference on Computer Vision
Approximating maximum clique with a Hopfield network
IEEE Transactions on Neural Networks
| Hi-index | 0.00 |
Our aim is to provide an autonomous vehicle moving into an indoorenvironment with a visual system to perform a qualitative 3D structurereconstruction of the surrounding environment by recovering the differentplanar surfaces present in the observed scene.The method is based on qualitative detection of planar surfacesby using projective invariant constraints without the use of depth estimates.The goal is achieved by analyzing two images acquired by observingthe scene from two different points of view. The method can beapplied to both stereo images and motion images.Our method recovers planar surfaces by clustering high variance interestpoints whose cross ratio measurements are preserved in two differentperspective projections. Once interest points are extracted from eachimage, the clusteringprocess requires to grouping corresponding points by preserving thecross ratio measurements.We solve the twofold problemof finding corresponding points and grouping the coplanar onesthrough a global optimization approach based onmatching of high relational graphs and clustering on thecorresponding association graph through a relaxation labeling algorithm.Through our experimental tests, we found themethod to be very fast to converge to a solution, showing howhigher order interactions, instead to giving rise to a more complexproblem, help to speed-up the optimization process and to reach atsame time good results.