Presentation 2001/10/11
Fractal-based Image Disocclusion using Kernel Principal Components Analysis
Shintaro IZOE, Yukiko KENMOCHI, Kazunori KOTANI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We provide an automated method to repair broken, occluded images. We call this method an image disocclusion. For estimation of the original image occluded by objects, our disocclusion method exploits fractal and locality of images. Also it operates local images in the whole image as vectors. First, our method obtains some local images which is similar to the occluded image. Using a subspace spanned by these image vectors, our method estimates original image. Accuracy of image estimation also depends on dimension of the subspace. Keeping this in mind, we will next propose the new method of image disocclusion using kernel principal components analysis.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image Disocclusion / Kernel Principal Components Analysis / Fractal
Paper # PRMU2001-106,NC2001-56
Date of Issue

Conference Information
Committee PRMU
Conference Date 2001/10/11(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Fractal-based Image Disocclusion using Kernel Principal Components Analysis
Sub Title (in English)
Keyword(1) Image Disocclusion
Keyword(2) Kernel Principal Components Analysis
Keyword(3) Fractal
1st Author's Name Shintaro IZOE
1st Author's Affiliation Graduate School of Information Science, Japan Advanced Institute of Science and Technology()
2nd Author's Name Yukiko KENMOCHI
2nd Author's Affiliation Graduate School of Information Science, Japan Advanced Institute of Science and Technology
3rd Author's Name Kazunori KOTANI
3rd Author's Affiliation Graduate School of Information Science, Japan Advanced Institute of Science and Technology
Date 2001/10/11
Paper # PRMU2001-106,NC2001-56
Volume (vol) vol.101
Number (no) 362
Page pp.pp.-
#Pages 8
Date of Issue