An Application of Markov Random Fields to Range Sensing

By James Diebel, Sebastian Thrun. Neural Information Processing Systems, 2005

704 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/807b81e5ab9fe35e4fdadc39913b2938ef553e8a

How researchers see this paper

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

Citations in the corpus

Beat the MTurkers: Automatic Image Labeling from Weak 3D Supervision (2014, methodology): “Note that this energy generalizes [5] to p-norms: When p = 2 we obtain a Gaussian MRF, and when p = 1 a Laplacian MRF.”
Depth Map Super-Resolution by Deep Multi-Scale Guidance (2016, methodology): “We do not explicitly formulate an optimization problem as the global methods [2, 4, 5, 15] or design a fixed filter as the local methods [6, 13, 14] because CNN can be trained to address the upsampling problem.”
Deep Joint Image Filtering (2016, methodology): “Among them, JBU [4], GF [6] and Ham [13] are generic methods for joint image upsampling while MRF [16], TGV [3] and Park [2] are algorithms specifically designed for image guided depth upsampling.”
Spatial-Depth Super Resolution for Range Images (2007, methodology): “Most related to our work is a range-enhanced method by Diebel and Thrun [8], in which a Markov Random Field (MRF) is ﬁrst designed based on the low resolution depth maps and high resolution camera images.”
The Fast Bilateral Solver (2015, methodology): “[7], which we ran ourselves using the publicly available code 11 , with the code was modiﬁed to use non-log color reference images.”
Sparsity Invariant CNNs (2017, background): “More advanced approaches are based on global energy minimization [5, 40, 9, 1, 42], compressive sensing [18], or incorporate semantics for improved performance [49].”
Depth and surface normal estimation from monocular images using regression on deep features and hierarchical CRFs (2015, methodology): “Potential 3: Auto-regression model Here we use the auto-regression model to characterize the local correlation structure in the depth map, which has been used in image colorization [25], depth in-painting [4], and depth image super resolution [26, 27].”
Image Guided Depth Upsampling Using Anisotropic Total Generalized Variation (2013, methodology): “Diebel and Thrun [5] performed an upsampling using a MRF formulation, where the smoothness term is weighted according to texture derivatives.”
Deep Depth Completion of a Single RGB-D Image (2018, background): “They have exploited a variety of approaches, including Markov random fields [44, 12, 42, 51, 58], shape-from-shading [23, 71], segmentation [41], and dictionary methods [18, 30, 45, 63].”
On the segmentation of 3D LIDAR point clouds (2011, methodology): “The former up-sample range data using the technique in [3] to”
3D point cloud segmentation: A survey (2013, methodology): “This method uses Markov Random Field [3] to estimate a 3D point corresponding to each image pixel.”
High quality depth map upsampling for 3D-TOF cameras (2011, methodology): “We compare our results with bilinear interpolation, MRF [6], bilateral ﬁlter [19], and a recent work on guided image ﬁlter [10].”
Self-Supervised Sparse-to-Dense: Self-Supervised Depth Completion from LiDAR and Monocular Camera (2018, background): “As a side note, the inpainting problem also finds close connection to depth denoising [13] and depth super-resolution [14–19].”
Color-Guided Depth Recovery From RGB-D Data Using an Adaptive Autoregressive Model (2014, methodology): “Our method is to recover high quality depth maps from low quality observations, and all the four kinds of degradations are handled in the proposed uniﬁed depth recovery framework.”