Sparsifying orthogonal transforms with compact bases for data compression

Junhui Hou; Hui Liu; Lap-Pui Chau; Ying He; Jie Chen

doi:10.1109/APSIPA.2016.7820905

Sparsifying orthogonal transforms with compact bases for data compression

Junhui Hou, Hui Liu, Lap-Pui Chau, Ying He, Jie Chen

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

1 Citation (Scopus)

Abstract

Learned Sparsifying orthogonal transforms (SOTs) have proven to be a powerful tool for image and video processing. In this paper, we propose a variant of SOT, named compact bases SOT, or CB-SOT, which has several promising features for data compression: (i) as an input-adaptive transform, it can sparsely represent the input data very well; (ii) the transform matrix is orthogonal; (iii) unlike SOT, the transform matrix is compact, since a large amount of entries are zero. We formulate CB-SOT as a constrained optimization problem and solve it efficiently using alternating iteration. Experiments on images show that the proposed algorithm empirically converges well and CB-SOT produces better performance of energy compaction, indicating its potential for data compression.

Original language	English
Title of host publication	2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016
Publisher	IEEE
ISBN (Print)	9789881476821
DOIs	https://doi.org/10.1109/APSIPA.2016.7820905
Publication status	Published - Dec 2016
Externally published	Yes
Event	2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016 - Jeju, Korea, Republic of Duration: 13 Dec 2016 → 16 Dec 2016

Conference

Conference	2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016
Place	Korea, Republic of
City	Jeju
Period	13/12/16 → 16/12/16

Research Keywords

image compression
Nonlinear approximation
optimization
orthogonal transform
sparse representation

Access Output

10.1109/APSIPA.2016.7820905

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{5a94da70ad3f44dc8c3b4305e5231161,

title = "Sparsifying orthogonal transforms with compact bases for data compression",

abstract = "Learned Sparsifying orthogonal transforms (SOTs) have proven to be a powerful tool for image and video processing. In this paper, we propose a variant of SOT, named compact bases SOT, or CB-SOT, which has several promising features for data compression: (i) as an input-adaptive transform, it can sparsely represent the input data very well; (ii) the transform matrix is orthogonal; (iii) unlike SOT, the transform matrix is compact, since a large amount of entries are zero. We formulate CB-SOT as a constrained optimization problem and solve it efficiently using alternating iteration. Experiments on images show that the proposed algorithm empirically converges well and CB-SOT produces better performance of energy compaction, indicating its potential for data compression.",

keywords = "image compression, Nonlinear approximation, optimization, orthogonal transform, sparse representation",

author = "Junhui Hou and Hui Liu and Lap-Pui Chau and Ying He and Jie Chen",

year = "2016",

month = dec,

doi = "10.1109/APSIPA.2016.7820905",

language = "English",

isbn = "9789881476821",

booktitle = "2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016",

publisher = "IEEE",

address = "United States",

note = "2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016 ; Conference date: 13-12-2016 Through 16-12-2016",

}

Hou, J, Liu, H, Chau, L-P, He, Y & Chen, J 2016, Sparsifying orthogonal transforms with compact bases for data compression. in 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016., 7820905, IEEE, 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016, Jeju, Korea, Republic of, 13/12/16. https://doi.org/10.1109/APSIPA.2016.7820905

Sparsifying orthogonal transforms with compact bases for data compression. / Hou, Junhui; Liu, Hui; Chau, Lap-Pui et al.
2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016. IEEE, 2016. 7820905.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Sparsifying orthogonal transforms with compact bases for data compression

AU - Hou, Junhui

AU - Liu, Hui

AU - Chau, Lap-Pui

AU - He, Ying

AU - Chen, Jie

PY - 2016/12

Y1 - 2016/12

N2 - Learned Sparsifying orthogonal transforms (SOTs) have proven to be a powerful tool for image and video processing. In this paper, we propose a variant of SOT, named compact bases SOT, or CB-SOT, which has several promising features for data compression: (i) as an input-adaptive transform, it can sparsely represent the input data very well; (ii) the transform matrix is orthogonal; (iii) unlike SOT, the transform matrix is compact, since a large amount of entries are zero. We formulate CB-SOT as a constrained optimization problem and solve it efficiently using alternating iteration. Experiments on images show that the proposed algorithm empirically converges well and CB-SOT produces better performance of energy compaction, indicating its potential for data compression.

AB - Learned Sparsifying orthogonal transforms (SOTs) have proven to be a powerful tool for image and video processing. In this paper, we propose a variant of SOT, named compact bases SOT, or CB-SOT, which has several promising features for data compression: (i) as an input-adaptive transform, it can sparsely represent the input data very well; (ii) the transform matrix is orthogonal; (iii) unlike SOT, the transform matrix is compact, since a large amount of entries are zero. We formulate CB-SOT as a constrained optimization problem and solve it efficiently using alternating iteration. Experiments on images show that the proposed algorithm empirically converges well and CB-SOT produces better performance of energy compaction, indicating its potential for data compression.

KW - image compression

KW - Nonlinear approximation

KW - optimization

KW - orthogonal transform

KW - sparse representation

UR - http://www.scopus.com/inward/record.url?scp=85013809585&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85013809585&origin=recordpage

U2 - 10.1109/APSIPA.2016.7820905

DO - 10.1109/APSIPA.2016.7820905

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9789881476821

BT - 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016

PB - IEEE

T2 - 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016

Y2 - 13 December 2016 through 16 December 2016

ER -

Sparsifying orthogonal transforms with compact bases for data compression

Abstract

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this