Please use this identifier to cite or link to this item: http://hdl.handle.net/2440/107516
Citations
Scopus Web of Science® Altmetric
?
?
Type: Conference paper
Title: Efficient SDP inference for fully-connected CRFs based on low-rank decomposition
Author: Wang, P.
Shen, C.
Van Den Hengel, A.
Citation: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2015 / vol.07-12-June-2015, pp.3222-3231
Publisher: IEEE
Issue Date: 2015
Series/Report no.: IEEE Conference on Computer Vision and Pattern Recognition
ISBN: 9781467369640
ISSN: 1063-6919
Conference Name: Conference on Computer Vision and Pattern Recognition (CVPR) (07 Jun 2015 - 12 Jun 2015 : Boston, MA)
Statement of
Responsibility: 
Peng Wang, Chunhua Shen, Anton van den Hengel
Abstract: Conditional Random Fields (CRFs) are one of the core technologies in computer vision, and have been applied to a wide variety of tasks. Conventional CRFs typically define edges between neighboring image pixels, resulting in a sparse graph over which inference can be performed efficiently. However, these CRFs fail to model more complex priors such as long-range contextual relationships. Fullyconnected CRFs have thus been proposed. While there are efficient approximate inference methods for such CRFs, usually they are sensitive to initialization and make strong assumptions. In this work, we develop an efficient, yet general SDP algorithm for inference on fully-connected CRFs. The core of the proposed algorithm is a tailored quasi- Newton method, which solves a specialized SDP dual problem and takes advantage of the low-rank matrix approximation for fast computation. Experiments demonstrate that our method can be applied to fully-connected CRFs that could not previously be solved, such as those arising in pixel-level image co-segmentation.
Rights: Copyright © 2015, IEEE
RMID: 0030046290
DOI: 10.1109/CVPR.2015.7298942
Appears in Collections:Computer Science publications

Files in This Item:
File Description SizeFormat 
RA_hdl_107516.pdfRestricted Access339.07 kBAdobe PDFView/Open


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