Products of Edge-perts

Peter V. Gehler, Max Welling

Abstract

Images represent an important and abundant source of data. Understanding their statistical structure has important applications such as image compression and restoration. In this paper we propose a particular kind of probabilistic model, dubbed the "products of edge-perts model" to describe the structure of wavelet transformed images. We develop a practical denoising algorithm based on a single edge-pert and show state-ofthe-art denoising performance on benchmark images.
BibTeX
@inproceedings{gehler2005products,
  author = {Peter V. Gehler and Max Welling},
  title = {{Products of Edge-perts}},
  booktitle = {Advances in Neural Information Processing Systems},
  year = {2005},
}
Figure