Global stereo reconstruction under second order smoothness priors

By Oliver J. Woodford, Philip H. S. Torr, Philip H. S. Torr, Ian David Reid, Andrew William Fitzgibbon. 2008

393 citations recorded by the source corpus.

Subfields: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

Semi-global Stereo Matching with Surface Orientation Priors (2017, background): “[47] show how secondorder smoothness terms can be efficiently optimized via QBPO.”
Occlusion-Aware Depth Estimation Using Light-Field Cameras (2015, methodology): “[25] imposed an additional second order smoothness term in the optimization, and solved it using Quadratic Pseudo-Boolean Optimization [19].”
The structure-from-motion reconstruction pipeline - a survey with focus on short image sequences (2010, background): “One approach to create a dense surface is to apply a disparity algorithm [5, 10, 18, 48] on a rectified image pair [19, 32].”
A category-level 3-D object dataset: Putting the Kinect to work (2011, background): “We assume that objects have second order smoothness (that curvature is minimized)—a classic prior on natural shapes [16, 26].”
Joint motion estimation and segmentation of complex scenes with label costs and occlusion modeling (2012, methodology): “Results are shown on several classical motion segmentation and optical flow examples.”
Displets: Resolving stereo ambiguities using object knowledge (2015, methodology): “As firstorder approaches such as SGM are not able to express planarity priors, second-order models have recently been investigated either by introducing triple cliques into the formulation [63], by increasing the parameter space [34,43] or by following a total generalized variational (TGV) approach ”
Deformable Medical Image Registration: A Survey (2013, methodology): “Several variants of these methods, like fusion moves [393] or dynamic graph-cuts [394] were introduced to further improve convergence at the expense of becoming problem speci c.”
Multi-View Stereo: A Tutorial (2015, background): “5, the second order smoothness prior by Woodford, Torr, Reid, and Fitzgibbon is a natural extension to this, which favors piecewise planar surfaces [196].”
Harmony Potentials (2011, methodology): “In (Winn and Jojic, 2005), an unsupervised procedure is able to segment an object class using a learned class mask and a deformation field.”
Higher-order clique reduction in binary graph cut (2009, methodology): “[5]), [10] Second → 1st order [14, 8] → binary (fusion [29]) Higher → 1st order (this paper) → binary (fusion, this paper)”
MAP Estimation of Semi-Metric MRFs via Hierarchical Graph Cuts (2009, methodology): “the st-MINCUT algorithm, e.g. in αβ-swap [4] andα-expansion [5]. However, recently researchers have also used more sophisticated algorithms such as quadratic pseudo-boolean optimization [2] (e.g. see [13, 21, 35]). Move making algorithms are generally preferred in applications which involve a large ”
Cross-Based Local Stereo Matching Using Orthogonal Integral Images (2009, methodology): “It even achieves a better performance than some complex global stereo matching methods [16]–[18], as shown in Table III.”
Shape, albedo, and illumination from a single image of an unknown object (2012, background): “In comparison, the Laplacian of Z and the second partial derivatives of Z can be made large simply due to foreshortening, which means that priors placed on these quantities [2, 29] would prefer certain shapes simply due to the angle from which those shapes are observed — clearly undesirable.”
Deep Virtual Stereo Odometry: Leveraging Deep Depth Prediction for Monocular Direct Sparse Odometry (2018, methodology): “We use a second-order smoothness prior [43] and downweight it when the image gradient is high [18].”
Efficient Large-Scale Stereo Matching (2010, methodology): “Higher order cliques can overcome these problems [7], but they are even more computationally demanding.”
High-quality single-shot capture of facial geometry (2010, methodology): “Following [Woodford et al. 2008] we employ second-order properties, but use them”
Deformable medical image registration: setting the state of the art with discrete methods. (2011, neutral): “Computational performance (70, 71, 73, 74) is the main strength of these methods while constraints (72) on the type of energies that can be minimized is their main bottleneck.”
PatchMatch Stereo - Stereo Matching with Slanted Support Windows (2011, methodology): “As a smoothness term, we use the second-order term of [13], which overcomes the fronto-parallel bias of competing terms (e.”