Nonparametric image interpolation and dictionary learning using spatially-dependent dirichlet and beta process priors

We present a Bayesian model for image interpolation and dictionary learning that uses two nonparametric priors for sparse signal representations: the beta process and the Dirichlet process. Additionally, the model uses spatial information within the image to encourage sharing of information within image subregions. We derive a hybrid MAP/Gibbs sampler, which performs Gibbs sampling for the latent indicator variables and MAP estimation for all other parameters. We present experimental results, where we show an improvement over other state-of-the-art algorithms in the low-measurement regime. © 2010 IEEE.

Full Text

Duke Authors

Cited Authors

  • Paisley, J; Zhou, M; Sapiro, G; Carin, L

Published Date

  • 2010

Published In

Start / End Page

  • 1869 - 1872

International Standard Serial Number (ISSN)

  • 1522-4880

Digital Object Identifier (DOI)

  • 10.1109/ICIP.2010.5653350