Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Conference paper
Title: An adversarial optimization approach to efficient outlier removal
Author: Yu, J.
Eriksson, A.
Chin, T.
Suter, D.
Citation: 2011 IEEE International Conference on Computer Vision, 2011: pp.399-406
Publisher: IEEE
Publisher Place: 345 E 47TH ST, NEW YORK, NY 10017 USA
Issue Date: 2011
Series/Report no.: IEEE International Conference on Computer Vision
ISBN: 9781457711015
ISSN: 1550-5499
Conference Name: International Conference on Computer Vision (13th : 2011 : Barcelona, Spain)
Statement of
Jin Yu, Anders Eriksson, Tat-Jun Chin, David Suter
Abstract: This paper proposes a novel adversarial optimization approach to efficient outlier removal in computer vision. We characterize the outlier removal problem as a game that involves two players of conflicting interests, namely, optimizer and outlier. Such an adversarial view not only brings new insights into various existing methods, but also gives rise to a general optimization framework that provably unifies them. Under the proposed framework, we develop a new outlier removal approach that is able to offer a much needed control over the trade-off between reliability and speed, which is otherwise not available in previous methods. The proposed approach is driven by a mixed-integer minmax (convex-concave) optimization process. Although a minmax problem is generally not amenable to efficient optimization, we show that for some commonly used vision objective functions, an equivalent Linear Program reformulation exists. We demonstrate our method on two representative multiview geometry problems. Experiments on real image data illustrate superior practical performance of our method over recent techniques.
Rights: Copyright © 2011 by IEEE.
RMID: 0020116865
DOI: 10.1109/ICCV.2011.6126268
Description (link):
Published version:
Appears in Collections:Computer Science publications

Files in This Item:
File Description SizeFormat 
RA_hdl_70329.pdfRestricted Access4.02 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.