Superpixel-driven graph transform for image compression

Giulia Fracastoro, Francesco Verdoja, Marco Grangetto, Enrico Magli

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

19 Citations (Scopus)


Block-based compression tends to be inefficient when blocks contain arbitrary shaped discontinuities. Recently, graph-based approaches have been proposed to address this issue, but the cost of transmitting graph topology often overcome the gain of such techniques. In this work we propose a new Superpixel-driven Graph Transform (SDGT) that uses clusters of superpixels, which have the ability to adhere nicely to edges in the image, as coding blocks and computes inside these homogeneously colored regions a graph transform which is shape-adaptive. Doing so, only the borders of the regions and the transform coefficients need to be transmitted, in place of all the structure of the graph. The proposed method is finally compared to DCT and the experimental results show how it is able to outperform DCT both visually and in term of PSNR.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings
PublisherIEEE Computer Society
Number of pages5
ISBN (Electronic)9781479983391
Publication statusPublished - 9 Dec 2015
MoE publication typeA4 Article in a conference publication
EventIEEE International Conference on Image Processing - Quebec City, Canada
Duration: 27 Sep 201530 Sep 2015


ConferenceIEEE International Conference on Image Processing
Abbreviated titleICIP
CityQuebec City


  • clustering
  • graph transform
  • Image compression
  • superpixels


Dive into the research topics of 'Superpixel-driven graph transform for image compression'. Together they form a unique fingerprint.
  • Best 10% Paper

    Verdoja, Francesco (Recipient), Sep 2015

    Prize: Award or honor granted for a specific work

Cite this