Presentation | 2005/6/16 Super-resolution with Smooth-Gap Prior Atsunori KANEMURA, Shinichi MAEDA, Shin ISHII, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | When extracting a single high resolution image from a set of low resolution images, maximum a posteiori estimators have been recongnized to be effective because they resolve the ill-posed nature of this inverse problem. A Bayesian approach which marginalizes over the high resolution image and then estimates the degrading process has also been established recently. However, the formerly used prior distribution was a simple Gaussian distribution, which prefers blurred high resolution images. In this study we use a smooth-gap prior which describes inherent properties of images more faithfully in order to estimate both the degrading process and the high resolution image. The smooth-gap prior is a complicated Gaussian distribution but allows estimated images to be sharp. Under the non-blind condition (in which we know the degrading process) the smooth-gap prior worked effectively to make estimated high resolution images sharp. Under the blind condition (in which we do not know the degrading process) the smooth-gap prior was comparable to the simple Gaussian prior in the estimation of the degrading process whereas superior to the Gaussian in respect of the sharpness of the image. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Super-resolution / blind deconvolution / Bayesian inference / image processing / prior distribution |
Paper # | NC2005-13 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2005/6/16(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Neurocomputing (NC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Super-resolution with Smooth-Gap Prior |
Sub Title (in English) | |
Keyword(1) | Super-resolution |
Keyword(2) | blind deconvolution |
Keyword(3) | Bayesian inference |
Keyword(4) | image processing |
Keyword(5) | prior distribution |
1st Author's Name | Atsunori KANEMURA |
1st Author's Affiliation | Graduate School of Information Science, Nara Institute of Science and Technology() |
2nd Author's Name | Shinichi MAEDA |
2nd Author's Affiliation | Graduate School of Information Science, Nara Institute of Science and Technology |
3rd Author's Name | Shin ISHII |
3rd Author's Affiliation | Graduate School of Information Science, Nara Institute of Science and Technology |
Date | 2005/6/16 |
Paper # | NC2005-13 |
Volume (vol) | vol.105 |
Number (no) | 130 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |