A common framework for image segmentation

By Davi Geiger, Alan L. Yuille. 1990

Subfields: 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 Vision and Pattern Recognition

Citations in the corpus

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs (2016, background): “We note that mean field inference had been extensively studied for traditional image segmentation tasks [54], [55], [56], but these older models were typically limited to short-range connections.”
Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs (2014, background): “We note that mean ﬁeld had been extensively studied for traditional image segmentation/edge detection tasks, e.g ., (Geiger & Girosi, 1991; Geiger & Yuille, 1991; Kokkinos et al., 2008), but recently Kr ¨ ahenb ¨ uhl & Koltun (2011) showed that the inference can be very efﬁcient for fully connected…”
On the Link between Gaussian Homotopy Continuation and Convex Envelopes (2015, methodology): “Motivated by problems in stereo and template matching, Yuille popularized MFA in a series of works [30,74,75,76,77,78].”
SoftPOSIT: Simultaneous Pose and Correspondence Determination (2002, methodology): “This is a deterministic annealing process (Geiger and Yuille, 1991) known as softmax (Bridle, 1990).”
Robust Point Matching via Vector Field Consensus (2014, methodology): “We also show that in the special case where there is an underlying parametric geometrical model (e.g., the epipolar line constraint) that we obtain better results than standard alternatives like RANSAC if a large number of outliers are present.”
Geodesic Active Regions: A New Framework to Deal with Frame Partition Problems in Computer Vision (2002, methodology): “The region-based methods are more suitable approaches to deal with frame partition problems and can be roughly classi ed into two categories: The regiongrowing techniques [1, 26] and the Markov Random Fields based approaches [2, 9, 10].”
Prior Learning and Gibbs Reaction-Diffusion (1997, methodology): “These prior models have been motivated by regularization theory [26], [18],(1) physical modeling [31], [4],(2) Bayesian theory [9], [20], and robust statistics [19], [13], [3].”
Computational analysis and learning for a biologically motivated model of boundary detection (2008, methodology): “This means that solving the system of ODE's determining the activity of Stage II neurons amounts to performing the MFA on a Boltzmann Machine.”
Structured Learning and Prediction in Computer Vision (2011, methodology): “It has been used parameter estimation in MRF in the early 1990s, [48], and more recently to learn the parameters of CRF in image segmentation, [159], and stereo depth estimation, [161].”
Synergism in low level vision (2002, unknown): “Information about homogeneity and discontinuity can be also captured by statistical measures and fused under Bayesian rules, e.g., [6], [7].”