Skip to main content

A Bayesian Nonparametric Approach to Image Super-Resolution.

Author(s): Polatkan, Gungor; Zhou, Mingyuan; Carin, Lawrence; Blei, David; Daubechies, Ingrid

To refer to this page use:
Abstract: Super-resolution methods form high-resolution images from low-resolution images. In this paper, we develop a new Bayesian nonparametric model for super-resolution. Our method uses a beta-Bernoulli process to learn a set of recurring visual patterns, called dictionary elements, from the data. Because it is nonparametric, the number of elements found is also determined from the data. We test the results on both benchmark and natural images, comparing with several other models from the research literature. We perform large-scale human evaluation experiments to assess the visual quality of the results. In a first implementation, we use Gibbs sampling to approximate the posterior. However, this algorithm is not feasible for large-scale data. To circumvent this, we then develop an online variational Bayes (VB) algorithm. This algorithm finds high quality dictionaries in a fraction of the time needed by the Gibbs sampler.
Publication Date: 1-Feb-2015
Citation: Polatkan, Gungor, Zhou, Mingyuan, Carin, Lawrence, Blei, David, Daubechies, Ingrid. (2015). A Bayesian Nonparametric Approach to Image Super-Resolution.. IEEE transactions on pattern analysis and machine intelligence, 37 (2), 346 - 358. doi:10.1109/tpami.2014.2321404
DOI: doi:10.1109/tpami.2014.2321404
ISSN: 0162-8828
EISSN: 1939-3539
Pages: 346 - 358
Language: eng
Type of Material: Journal Article
Journal/Proceeding Title: IEEE Transactions on Pattern Analysis and Machine Intelligence
Version: Author's manuscript

Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.