Global Stereo Reconstruction under Second-Order Smoothness Priors

Authors:
Oliver Woodford;Philip Torr;Ian Reid;Andrew Fitzgibbon
Affiliations:
University of Oxford, Oxford;Oxford Brookes University, Oxford;University of Oxford, Oxford;Microsoft Research Ltd., Cambridge
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
2009

Citing 0
Cited 17

Efficient large-scale stereo matching

ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part I
Triangulating a plane

SCIA'11 Proceedings of the 17th Scandinavian conference on Image analysis
Multiview face capture using polarized spherical gradient illumination

Proceedings of the 2011 SIGGRAPH Asia Conference
Curvature regularization for curves and surfaces in a global optimization framework

EMMCVPR'11 Proceedings of the 8th international conference on Energy minimization methods in computer vision and pattern recognition
Harmony Potentials

International Journal of Computer Vision
An optimal time---space algorithm for dense stereo matching

Journal of Real-Time Image Processing
Generalized roof duality

Discrete Applied Mathematics
Approximate MRF inference using bounded treewidth subgraphs

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part I
Filter-Based mean-field inference for random fields with higher-order terms and product label-spaces

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part V
Contraction moves for geometric model fitting

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part VII
Tighter relaxations for higher-order models based on generalized roof duality

ECCV'12 Proceedings of the 12th international conference on Computer Vision - Volume Part III
Window annealing for pixel-labeling problems

Computer Vision and Image Understanding
Multi-view Scene Flow Estimation: A View Centered Variational Approach

International Journal of Computer Vision
Stereo reconstruction and contrast restoration in daytime fog

ACCV'12 Proceedings of the 11th Asian conference on Computer Vision - Volume Part IV
Markov Random Field modeling, inference & learning in computer vision & image understanding: A survey

Computer Vision and Image Understanding
Quadratic Transformation for Planar Mapping of Implicit Surfaces

Journal of Mathematical Imaging and Vision
A homography transform based higher-order MRF model for stereo matching

Pattern Recognition Letters

Quantified Score

Hi-index	0.14

Visualization

Abstract

Second-order priors on the smoothness of 3D surfaces are a better model of typical scenes than first-order priors. However, stereo reconstruction using global inference algorithms, such as graph cuts, has not been able to incorporate second-order priors because the triple cliques needed to express them yield intractable (nonsubmodular) optimization problems. This paper shows that inference with triple cliques can be effectively performed. Our optimization strategy is a development of recent extensions to \alpha--expansion, based on the “ QPBO” algorithm. The strategy is to repeatedly merge proposal depth maps using a novel extension of QPBO. Proposal depth maps can come from any source, for example, frontoparallel planes as in \alpha-expansion, or indeed any existing stereo algorithm, with arbitrary parameter settings.