Learning Conditional Random Fields for Stereo

By Daniel Scharstein, C. Pal. 2007

1,009 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Probabilistic & Bayesian Methods, 3D Vision & Pose

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Probabilistic & Bayesian Methods

In 3D Vision & Pose

Citations in the corpus

Structured Convolutional Kernel Networks for Airline Crew Scheduling (2021, methodology): “We ﬁrst run an approximate MAP inference algorithm and then imagine that the marginals put a unit probability at the approximate MAP solution and zero elsewhere (Kumar et al., 2005; Zhong & Wang, 2007; Scharstein & Pal, 2007).”
IRS: A Large Synthetic Indoor Robotics Stereo Dataset for Disparity and Surface Normal Estimation (2019, background): “Existing studies [10], [11], [12], [13] proposed stereo datasets collected by real sensing hardware, which contributed a lot to stereo vision research.”
Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches (2015, methodology): “The data sets were published in ﬁve separate works in the years 2001, 2003, 2005, 2006, and 2014 (Scharstein and Szeliski, 2002, 2003; Scharstein and Pal, 2007; Hirschm¨uller and Scharstein, 2007; Scharstein et al., 2014).”
Image Guided Depth Upsampling Using Anisotropic Total Generalized Variation (2013, methodology): “An exhaustive evaluation of our method in terms of quantitative and qualitative comparison is made using input images from the Middlebury datasets [10] [20].”
Global stereo reconstruction under second order smoothness priors (2008, methodology): “The commonest image feature used in W (·) is the magnitude of the local image gradient of that neighbourhood, parallel to the neighbourhood [10], [11], [14], [21], [60], [61], [62].”
Learning for Disparity Estimation Through Feature Constancy (2017, neutral): “The four-step pipeline dominants existing stereo matching algorithms [24, 22, 9, 21], while each of its steps is important to the overall stereo matching performance.”
Computing the stereo matching cost with a convolutional neural network (2014, methodology): “[13] constructed a new dataset of 30 stereo pairs and used it to learn parameters of a conditional random field.”
A Multi-view Stereo Benchmark with High-Resolution Images and Multi-camera Videos (2017, background): “Indeed, the advent of excellent datasets and benchmarks, such as [6,12,17,29,32–34,38,40], has greatly advanced the state of the art of stereo and multi-view stereo (MVS) techniques.”
DehazeNet: An End-to-End System for Single Image Haze Removal (2016, background): “There are 12 pairs of stereo images collected in Middlebury Stereo Datasets (2001-2006) [44], [45], [46].”
Learning Graphical Model Parameters with Approximate Marginal Inference (2013, neutral): “20 by running a (presumably approximate) MAP inference algorithm, and then imagining that the marginals put unit probability at the approximate MAP solution, and zero elsewhere [27], [28], [21].”
AOD-Net: All-in-One Dehazing Network (2017, methodology): “We also take the 800 full-size synthetic images from the Middlebury stereo database [19, 18, 9] as the TestSet B.”
A Database and Evaluation Methodology for Optical Flow (2007, positive): “Specifically we use the Venus dataset obtained by registering planes in the scene [19], and the Moebius dataset [18], which was obtained using structured lighting [20].”
Constant Time Weighted Median Filtering for Stereo Matching and Beyond (2013, methodology): “We evaluate all the pairs that have ground-truths available: the standard 4 pairs, the 2001 set (6 pairs), the 2005 set (6 pairs), and the 2006 set (21 pairs) [23, 12].”
Learning optimized MAP estimates in continuously-valued MRF models (2009, methodology): “In [12], Scharstein and Pal propose another method which uses MAP estimates to compute approximate expectation gradients to learn parameters of a CRF model.”
Who killed the directed model? (2008, positive): “These have been successfully learnt as priors for natural images [13] [22] [23], stereo depth [16] [15], single image depth [14], object segmentation [2], optical flow [12], intrinsic images [20], etc.”
Towards perceptual shared autonomy for robotic mobile manipulation (2011, background): “The first two images are courtesy of the Middlebury Stereo Dataset [23].”
End-to-End Learning of Geometry and Context for Deep Stereo Regression (2017, methodology): “Scharstein and Pal [38] learn conditional random ﬁeld (CRF) parameters, and Li and Huttenlocher [29] train a non-parametric CRF model using the structured support vector machine.”
LAF-Net: Locally Adaptive Fusion Networks for Stereo Confidence Estimation (2019, methodology): “The proposed method is extensively evaluated through an ablation study and comparison with conventional handcrafted and CNNs-based methods on various benchmarks, including Middlebury 2006 [34], Middlebury 2014 [33], and KITTI 2015 [24].”