Parallel lasso for large-scale video concept detection

Geng, B; Li, Y; Tao, D; Wang, M; Zha, ZJ; Xu, C

Parallel lasso for large-scale video concept detection

Geng, B Li, Y Tao, D

Wang, M Zha, ZJ Xu, C

Permalink

Publication Type:: Journal Article
Citation:: IEEE Transactions on Multimedia, 2012, 14 (1), pp. 55 - 65
Issue Date:: 2012-02-01

Closed Access

	Filename	Description	Size
	2012002987OK.pdf		1.19 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Geng, B	en_US
dc.contributor.author	Li, Y	en_US
dc.contributor.author	Tao, D https://orcid.org/0000-0001-7225-5449	en_US
dc.contributor.author	Wang, M	en_US
dc.contributor.author	Zha, ZJ	en_US
dc.contributor.author	Xu, C	en_US
dc.date.issued	2012-02-01	en_US
dc.identifier.citation	IEEE Transactions on Multimedia, 2012, 14 (1), pp. 55 - 65	en_US
dc.identifier.issn	1520-9210	en_US
dc.identifier.uri	http://hdl.handle.net/10453/22886
dc.description.abstract	Existing video concept detectors are generally built upon the kernel based machine learning techniques, e.g., support vector machines, regularized least squares, and logistic regression, just to name a few. However, in order to build robust detectors, the learning process suffers from the scalability issues including the high-dimensional multi-modality visual features and the large-scale keyframe examples. In this paper, we propose parallel lasso (Plasso) by introducing the parallel distributed computation to significantly improve the scalability of lasso (the regularized least squares). We apply the parallel incomplete Cholesky factorization to approximate the covariance statistics in the preprocess step, and the parallel primal-dual interior-point method with the Sherman-Morrison-Woodbury formula to optimize the model parameters. For a dataset with samples in a -dimensional space, compared with lasso, Plasso significantly reduces complexities from the original for computational time and for storage space to and respectively, if the system has $m$ processors and the reduced dimension is much smaller than the original dimension. Furthermore, we develop the kernel extension of the proposed linear algorithm with the sample reweighting schema, and we can achieve similar time and space complexity improvements [time complexity from to and the space complexity from to for a dataset with training examples]. Experimental results on TRECVID video concept detection challenges suggest that the proposed method can obtain significant time and space savings for training effective detectors with limited communication overhead. © 2006 IEEE.	en_US
dc.relation.ispartof	IEEE Transactions on Multimedia	en_US
dc.relation.isbasedon	10.1109/TMM.2011.2174781	en_US
dc.subject.classification	Artificial Intelligence & Image Processing	en_US
dc.title	Parallel lasso for large-scale video concept detection	en_US
dc.type	Journal Article
utslib.citation.volume	1	en_US
utslib.citation.volume	14	en_US
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
utslib.for	08 Information and Computing Sciences	en_US
utslib.for	09 Engineering	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	closed_access
pubs.issue	1	en_US
pubs.publication-status	Published	en_US
pubs.volume	14	en_US

Abstract:

Existing video concept detectors are generally built upon the kernel based machine learning techniques, e.g., support vector machines, regularized least squares, and logistic regression, just to name a few. However, in order to build robust detectors, the learning process suffers from the scalability issues including the high-dimensional multi-modality visual features and the large-scale keyframe examples. In this paper, we propose parallel lasso (Plasso) by introducing the parallel distributed computation to significantly improve the scalability of lasso (the regularized least squares). We apply the parallel incomplete Cholesky factorization to approximate the covariance statistics in the preprocess step, and the parallel primal-dual interior-point method with the Sherman-Morrison-Woodbury formula to optimize the model parameters. For a dataset with samples in a -dimensional space, compared with lasso, Plasso significantly reduces complexities from the original for computational time and for storage space to and respectively, if the system has $m$ processors and the reduced dimension is much smaller than the original dimension. Furthermore, we develop the kernel extension of the proposed linear algorithm with the sample reweighting schema, and we can achieve similar time and space complexity improvements [time complexity from to and the space complexity from to for a dataset with training examples]. Experimental results on TRECVID video concept detection challenges suggest that the proposed method can obtain significant time and space savings for training effective detectors with limited communication overhead. © 2006 IEEE.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/22886