Please use this identifier to cite or link to this item:
Title: Image decomposition with multilabel context: Algorithms and applications
Authors: Li, T.
Yan, S. 
Mei, T.
Hua, X.-S.
Kweon, I.-S.
Keywords: Image classification
image decomposition
multilabel context
Issue Date: Aug-2011
Citation: Li, T., Yan, S., Mei, T., Hua, X.-S., Kweon, I.-S. (2011-08). Image decomposition with multilabel context: Algorithms and applications. IEEE Transactions on Image Processing 20 (8) : 2301-2314. ScholarBank@NUS Repository.
Abstract: Most research on image decomposition, e.g., image segmentation and image parsing, has predominantly focused on the low-level visual clues within a single image and neglected the contextual information across images. In this paper, we present a new perspective to image decomposition piloted by the multilabel context associated with each individual image. Observing that the contextual information (i.e., local label representations of the same label are similar while those from different labels are dissimilar) exists across images, we propose to perform image decomposition in a collective way and obtain an optimal representation for each label from a set of multilabeled images. We formulate the problem as an optimization problem which maximizes inter-label difference while minimizing the intra-label difference of the target label representations and propose two ways to solve this problem. Such a contextual image decomposition has a wide variety of applications, among which two exemplary onesmultilabel image annotation and label ranking, are presented and evaluated with different classification techniques. Extensive experiments on two benchmark datasets demonstrate promising results. © 2010 IEEE.
Source Title: IEEE Transactions on Image Processing
ISSN: 10577149
DOI: 10.1109/TIP.2010.2103081
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.


checked on Mar 20, 2019


checked on Mar 20, 2019

Page view(s)

checked on Mar 9, 2019

Google ScholarTM



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