The v-Trick for Image Reconstruction

Rätsch, G; Tsuda, K

Local TagsRelease HistoryDetailsSummary

The v-Trick for Image Reconstruction

Rätsch, G., & Tsuda, K. (2004). The v-Trick for Image Reconstruction. In First Mathematical Programming Society International Conference on Continuous Optimization (ICCOPT I) (pp. 55).

Item is Released

show all hide all

Basic

show hide

Item Permalink: https://hdl.handle.net/21.11116/0000-0005-66FA-F Version Permalink: https://hdl.handle.net/21.11116/0000-0005-66FB-E

Genre: Meeting Abstract

Files

show Files

Locators

show

hide

Locator:
http://eaton.math.rpi.edu/iccopt/program.pdf (Abstract) Open Access status unknown

Description:
-

OA-Status:

Creators

show

hide

Creators:
Rätsch, G, Author
Tsuda, K^{1, 2}, Author

Affiliations:
1Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497794
2Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497795

Content

show

hide

Free keywords: -

Abstract: A common way of image denoising is to project a noisy image to the subspace ofadmissible images made for instance by PCA. However, a major drawback of thismethod is that all pixels are updated by the projection, even when only a few pixels arecorrupted by noise or occlusion. We propose a new method to identify the noisy pixelsby`1-norm penalization and update the identified pixels only. The identification andupdating of noisy pixels are formulated as a single linear program. The application ofthe well-knownν-trick is particularly useful in this context as it allows the specificationof the fraction of pixels to be updated. Moreover, we extend the linear program to beable to exploit prior knowledge that occlusions often appear in contiguous blocks (e.g.sunglasses on faces). The basic idea is to penalize boundary points and interior pointsof the occluded area differently. We are able to show theν-property also for thisextended LP leading a method which is easy to use. Experimental results impressivelydemonstrate the power of our approach.

Details

show

hide

Language(s):

Dates: Published Online: 2004-08

Publication Status: Published online

Pages: -

Publishing info: -

Table of Contents: -

Rev. Type: -

Identifiers: -

Degree: -

Event

show

hide

Title: First Mathematical Programming Society International Conference on Continuous Optimization (ICCOPT I)

Place of Event: Troy, NY, USA

Start-/End Date: 2004-08-02 - 2004-08-04

Legal Case

show

Project information

show

Source 1

show

hide

Title: First Mathematical Programming Society International Conference on Continuous Optimization (ICCOPT I)

Source Genre: Proceedings

Creator(s):

Affiliations:

Publ. Info: -

Pages: - Volume / Issue: - Sequence Number: MM 4 Start / End Page: 55 Identifier: -