Advanced search
1 file | 431.56 KB Add to list

Learning a piecewise linear transform coding scheme for images

Author
Organization
Abstract
Gaussian mixture models are among the most widely accepted methods for clustering and probability density estimation. Recently it has been shown that these statistical methods are perfectly suited for learning patch-based image priors for various image restoration problems. In this paper we investigate the use of GMM's for image compression. A piecewise linear transform coding scheme based on Vector Quantization is proposed. In this scheme two different learning algorithms for GMM's are considered and compared. Experimental results demonstrate that the proposed techniques outperform JPEG, with results comparable to JPEG2000 for a broad class of images.
Keywords
Image Compression, GMM's, MIXTURE-MODELS, VQ

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 431.56 KB

Citation

Please use this url to cite or link to this publication:

MLA
van den Oord, Aäron, Sander Dieleman, and Benjamin Schrauwen. “Learning a Piecewise Linear Transform Coding Scheme for Images.” Proceedings of SPIE, the International Society for Optical Engineering. Ed. Z Zhu. Vol. 8768. SPIE-INT SOC OPTICAL ENGINEERING, 2013. Print.
APA
van den Oord, A., Dieleman, S., & Schrauwen, B. (2013). Learning a piecewise linear transform coding scheme for images. In Z. Zhu (Ed.), Proceedings of SPIE, the International Society for Optical Engineering (Vol. 8768). Presented at the 4th International Conference on Graphic and Image Processing (ICGIP), SPIE-INT SOC OPTICAL ENGINEERING.
Chicago author-date
van den Oord, Aäron, Sander Dieleman, and Benjamin Schrauwen. 2013. “Learning a Piecewise Linear Transform Coding Scheme for Images.” In Proceedings of SPIE, the International Society for Optical Engineering, ed. Z Zhu. Vol. 8768. SPIE-INT SOC OPTICAL ENGINEERING.
Chicago author-date (all authors)
van den Oord, Aäron, Sander Dieleman, and Benjamin Schrauwen. 2013. “Learning a Piecewise Linear Transform Coding Scheme for Images.” In Proceedings of SPIE, the International Society for Optical Engineering, ed. Z Zhu. Vol. 8768. SPIE-INT SOC OPTICAL ENGINEERING.
Vancouver
1.
van den Oord A, Dieleman S, Schrauwen B. Learning a piecewise linear transform coding scheme for images. In: Zhu Z, editor. Proceedings of SPIE, the International Society for Optical Engineering. SPIE-INT SOC OPTICAL ENGINEERING; 2013.
IEEE
[1]
A. van den Oord, S. Dieleman, and B. Schrauwen, “Learning a piecewise linear transform coding scheme for images,” in Proceedings of SPIE, the International Society for Optical Engineering, Singapore, Singapore, 2013, vol. 8768.
@inproceedings{4088539,
  abstract     = {Gaussian mixture models are among the most widely accepted methods for clustering and probability density estimation. Recently it has been shown that these statistical methods are perfectly suited for learning patch-based image priors for various image restoration problems. In this paper we investigate the use of GMM's for image compression. A piecewise linear transform coding scheme based on Vector Quantization is proposed. In this scheme two different learning algorithms for GMM's are considered and compared. Experimental results demonstrate that the proposed techniques outperform JPEG, with results comparable to JPEG2000 for a broad class of images.},
  articleno    = {876844},
  author       = {van den Oord, Aäron and Dieleman, Sander and Schrauwen, Benjamin},
  booktitle    = {Proceedings of SPIE, the International Society for Optical Engineering},
  editor       = {Zhu, Z},
  isbn         = {9780819495662},
  issn         = {0277-786X},
  keywords     = {Image Compression,GMM's,MIXTURE-MODELS,VQ},
  language     = {eng},
  location     = {Singapore, Singapore},
  pages        = {5},
  publisher    = {SPIE-INT SOC OPTICAL ENGINEERING},
  title        = {Learning a piecewise linear transform coding scheme for images},
  url          = {http://dx.doi.org/10.1117/12.2011134},
  volume       = {8768},
  year         = {2013},
}

Altmetric
View in Altmetric
Web of Science
Times cited: