The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Dans cette lettre, nous avons appris les filtres surcomplets pour modéliser de riches a priori d'images de la nature. Notre approche étend les champs d'experts à mélange d'échelle gaussienne (GSM FOE), qui est un modèle approximatif rapide basé sur les champs d'experts (FOE). Dans ces modèles a priori d'images précédents, le cas surcomplet n'est pas pris en compte en raison de la lourdeur des calculs. Nous introduisons l'hypothèse de quasi-orthogonalité dans le GSM FOE, ce qui nous permet d'apprendre rapidement et efficacement des filtres surcomplets d'images de la nature. Les simulations montrent que les filtres surcomplets obtenus ont des propriétés similaires à celles de Fields of Experts, et les expériences de débruitage montrent également la supériorité de notre modèle.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Zhe WANG, Siwei LUO, Liang WANG, "A Fast Algorithm for Learning the Overcomplete Image Prior" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 2, pp. 403-406, February 2010, doi: 10.1587/transinf.E93.D.403.
Abstract: In this letter, we learned overcomplete filters to model rich priors of nature images. Our approach extends the Gaussian Scale Mixture Fields of Experts (GSM FOE), which is a fast approximate model based on Fields of Experts (FOE). In these previous image prior model, the overcomplete case is not considered because of the heavy computation. We introduce the assumption of quasi-orthogonality to the GSM FOE, which allows us to learn overcomplete filters of nature images fast and efficiently. Simulations show these obtained overcomplete filters have properties similar with those of Fields of Experts', and denoising experiments also show the superiority of our model.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.403/_p
Copier
@ARTICLE{e93-d_2_403,
author={Zhe WANG, Siwei LUO, Liang WANG, },
journal={IEICE TRANSACTIONS on Information},
title={A Fast Algorithm for Learning the Overcomplete Image Prior},
year={2010},
volume={E93-D},
number={2},
pages={403-406},
abstract={In this letter, we learned overcomplete filters to model rich priors of nature images. Our approach extends the Gaussian Scale Mixture Fields of Experts (GSM FOE), which is a fast approximate model based on Fields of Experts (FOE). In these previous image prior model, the overcomplete case is not considered because of the heavy computation. We introduce the assumption of quasi-orthogonality to the GSM FOE, which allows us to learn overcomplete filters of nature images fast and efficiently. Simulations show these obtained overcomplete filters have properties similar with those of Fields of Experts', and denoising experiments also show the superiority of our model.},
keywords={},
doi={10.1587/transinf.E93.D.403},
ISSN={1745-1361},
month={February},}
Copier
TY - JOUR
TI - A Fast Algorithm for Learning the Overcomplete Image Prior
T2 - IEICE TRANSACTIONS on Information
SP - 403
EP - 406
AU - Zhe WANG
AU - Siwei LUO
AU - Liang WANG
PY - 2010
DO - 10.1587/transinf.E93.D.403
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 2010
AB - In this letter, we learned overcomplete filters to model rich priors of nature images. Our approach extends the Gaussian Scale Mixture Fields of Experts (GSM FOE), which is a fast approximate model based on Fields of Experts (FOE). In these previous image prior model, the overcomplete case is not considered because of the heavy computation. We introduce the assumption of quasi-orthogonality to the GSM FOE, which allows us to learn overcomplete filters of nature images fast and efficiently. Simulations show these obtained overcomplete filters have properties similar with those of Fields of Experts', and denoising experiments also show the superiority of our model.
ER -