Fields of Experts: a framework for learning image priors

By Stefan Roth, Michael J. Black. 2005

1,236 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

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 Representation & Self-Supervised Learning

Citations in the corpus

A Theory of Generative ConvNet (2016, methodology): “The generative ConvNet model can be viewed as a hierarchical version of the FRAME model (Zhu et al., 1997), as well as the Product of Experts (Hinton, 2002; Teh et al., 2003) and Field of Experts (Roth & Black, 2005) models.”
GAN Inversion: A Survey (2021, methodology): “It has been widely adopted in computer vision tasks, including MRF [196], [197], [198], dark channel prior [199] and total variation regularizer [200], which all model correlation among neighboring pixels via Gibbs distribution.”
Generative Visual Manipulation on the Natural Image Manifold (2016, methodology): “Some methods attempt to capture full-image likelihoods [17] through dense patch overlap, though the basic building block is still small patches that do not capture global image structures and long range relations.”
Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks (2015, methodology): “For image processing tasks, models based on marginal distributions of image gradients are effective [20, 25], but are only designed for image restoration rather than being true density models (so cannot sample an actual image).”
Cooperative Training of Descriptor and Generator Networks (2016, methodology): “(a) The energy-based models [3], [4] or the Markov random field models [5], [6], where the probability distribution is defined by the feature statistics or the energy function computed from the image by a bottom-up process.”
Pixel Recursive Super Resolution (2017, background): “, 2×), super resolution techniques achieve satisfactory results [41, 8, 16, 39, 22] by building statistical prior models of images [35, 2, 51] that capture low-level characteristics of natural images.”
Internal statistics of a single natural image (2011, background): “, [16, 13, 15]) impose parametric distribution on natural image responses to local filters.”
Generating more realistic images using gated MRF's (2010, methodology): “These models are useful for extracting representations [2, 3, 4] that can be used for discriminative tasks and they can also provide adaptive priors [5, 6, 7] that can be used in applications like denoising and inpainting.”
Blind Motion Deblurring Using Image Statistics (2006, background): “Recent work in computer vision applied natural images priors for a variety of applications like denoising [25, 24], super resolution [27], video matting [2], inpainting [16] and reflections decomposition [15].”
Understanding the (un)interpretability of natural image distributions using generative models (2019, methodology): “[24] learns and applies image priors based on Fields of Experts.”
Natural image denoising: Optimality and inherent bounds (2011, methodology): “Image denoising algorithms have drastically advanced over the last few decades [17, 7, 20, 21, 3, 10, 8].”
Image Denoising Via Learned Dictionaries and Sparse representation (2006, methodology): “This aligns with a similar idea, appearing in [21], for turning a local MRF-based prior into a global one.”
Introspective Neural Networks for Generative Modeling (2017, background): “The history of generative modeling on image or nonimage domains is extremely rich, including the general image pattern theory [15], deformable models [48], inducing features [8], wake-sleep [19], the MiniMax entropy theory [51], the field of experts [36], Bayesian models [49], and deep belief nets [”
From Local Kernel to Nonlocal Multiple-Model Image Denoising (2009, methodology): “The Adaptive Principal Components algorithm by Muresan and Parks [78] and Fields of Experts (FoE) algorithm by Roth and Black [88], [89] are other successful examples of this sort of methods using bases optimized with respect to the given image or set of images at hand.”