Efficient Belief Propagation for Early Vision

By Pedro F. Felzenszwalb, Dan Huttenlocher. Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004., 2004

1,942 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/b3197ff5aa8f9cd36f98bcc8762b96250bdb4168

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

Citations in the corpus

2D Euclidean distance transform algorithms: A comparative survey (2008, methodology): “2004]; efficient belief-propagation [Felzenszwalb and Huttenlocher 2006]; multiphase flow simulations [Ceniceros and Roma 2005]; medical image analysis [Cuisenaire and Macq 1999b; hao Tseng et al.”
Symmetric stereo matching for occlusion handling (2005, methodology): “We adopted the implementation of belief propagation proposed by [8] for efficiency.”
Semantic Image Segmentation via Deep Parsing Network (2015, background): “Existing works such as [31, 29, 9, 34, 11, 2, 8, 25, 22] can be generally categorized into two groups based on their definitions of the unary and pairwise terms of MRF.”
SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, methodology): “In this paper, we will explore the SIFT flow algorithm in more depth and will demonstrate a wide array of applications for SIFT flow.”
Beyond pixels: exploring new representations and applications for motion analysis (2009, methodology): “To compute the message passing from pixel p to its neighbor q, we first gather all other messages and data term, and apply the routine in [36] to compute the message from p to q assuming that q and p have the same offset and range.”
A Tutorial on Dual Decomposition and Lagrangian Relaxation for Inference in Natural Language Processing (2012, methodology): “See the work of Lemaréchal (2001) and Fisher (1981) for surveys of LR methods, and the textbook of Korte and Vygen (2008) for background on combinatorial optimization.”
Efficient Large-Scale Stereo Matching (2010, methodology): “, graph cuts [4, 5] or belief propagation [6].”
Polonium: Tera-Scale Graph Mining and Inference for Malware Detection (2011, background): “The second optimization exploits the fact that the graph is bipartite (of machines and files) to reduce both the storage and computation for messages by half [29].”
Global stereo reconstruction under second order smoothness priors (2008, background): “A significant reason for this shift to first order priors lies in the development of powerful optimization techniques such as graph cuts [10], [16], [17], [18] and belief propagation (BP) [12], [ 13 ].”
SfM with MRFs: Discrete-Continuous Optimization for Large-Scale Structure from Motion (2013, neutral): “The innermost minimization can be performed for all lj in linear time using the generalized distance transform [39], while the other minimizations and the computation of Di can also be performed in linear time.”
Fast 3D Scanning with Automatic Motion Compensation (2007, methodology): “This is a labeling problem, which we formulate in an energy minimization framework and solve using Loopy Belief Propagation [6, 10].”
Plant classification system for crop /weed discrimination without segmentation (2014, methodology): “This cost function is minimized using efficient belief propagation [7] and the smoothed labeling l∗ is returned.”
Auto-tuning a high-level language targeted to GPU codes (2012, methodology): “We begin with the sequential code corresponding to the work by Felzenszwalb [12] available online.”
Distance Transforms of Sampled Functions (2012, methodology): “It has been used in computer vision for object recognition [11] and depth estimation [12] as well as in machine learning for maximum a posteriori inference [13, 19].”