Fields of experts: A framework for learning image priors

Fields of experts: A framework for learning image priors (2005)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Stefan Roth , Michael J. Black

Venue:	In CVPR
Citations:	292 - 4 self

BibTeX

@INPROCEEDINGS{Roth05fieldsof,
    author = {Stefan Roth and Michael J. Black},
    title = {Fields of experts: A framework for learning image priors},
    booktitle = {In CVPR},
    year = {2005}
}

OpenURL

Abstract

We develop a framework for learning generic, expressive image priors that capture the statistics of natural scenes and can be used for a variety of machine vision tasks. The approach extends traditional Markov Random Field (MRF) models by learning potential functions over extended pixel neighborhoods. Field potentials are modeled using a Products-of-Experts framework that exploits nonlinear functions of many linear filter responses. In contrast to previous MRF approaches all parameters, including the linear filters themselves, are learned from training data. We demonstrate the capabilities of this Field of Experts model with two example applications, image denoising and image inpainting, which are implemented using a simple, approximate inference scheme. While the model is trained on a generic image database and is not tuned toward a specific application, we obtain results that compete with and even outperform specialized techniques. 1.

Keyphrases

image prior field potential products-of-experts framework previous mrf expressive image prior potential function traditional markov random field linear filter many linear filter response approximate inference scheme extended pixel neighborhood machine vision task image denoising generic image database specific application nonlinear function expert model image inpainting example application natural scene