License: Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported license (CC BY-NC-ND 3.0)
When quoting this document, please refer to the following
DOI: 10.4230/OASIcs.MEMICS.2010.77
URN: urn:nbn:de:0030-drops-30681
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2011/3068/
Go to the corresponding OASIcs Volume Portal

Matela, Jiri ; Rusnak, Vit ; Holub, Petr

GPU-Based Sample-Parallel Context Modeling for EBCOT in JPEG2000

pdf-format:
12.pdf (0.5 MB)

Abstract

Embedded Block Coding with Optimal Truncation (EBCOT) is the fundamental and computationally very demanding part of the compression process of JPEG2000 image compression standard. EBCOT itself consists of two tiers. In Tier-1, image samples are compressed using context modeling and arithmetic coding. Resulting bit-stream is further formated and truncated in Tier-2. JPEG2000 has a number of applications in various fields where the processing speed and/or latency is a crucial attribute and the main limitation with state of the art implementations. In this paper we propose a new parallel approach to EBCOT context modeling that truly exploits massively parallel capabilities of modern GPUs and enables concurrent processing of individual image samples. Performance evaluation of our prototype shows speedup 12 times for the context modeller, and 1.4­5.3 times for the whole EBCOT Tier-1, which includes not yet optimized arithmetic coder.

BibTeX - Entry

@InProceedings{matela_et_al:OASIcs:2011:3068,
  author =	{Jiri Matela and Vit Rusnak and Petr Holub},
  title =	{{GPU-Based Sample-Parallel Context Modeling for EBCOT in JPEG2000}},
  booktitle =	{Sixth Doctoral Workshop on Mathematical and Engineering Methods in Computer Science (MEMICS'10) -- Selected Papers},
  pages =	{77--84},
  series =	{OpenAccess Series in Informatics (OASIcs)},
  ISBN =	{978-3-939897-22-4},
  ISSN =	{2190-6807},
  year =	{2011},
  volume =	{16},
  editor =	{Ludek Matyska and Michal Kozubek and Tom{\'a}{\v{s}} Vojnar and Pavel Zemc{\'i}k and David Antos},
  publisher =	{Schloss Dagstuhl--Leibniz-Zentrum fuer Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{http://drops.dagstuhl.de/opus/volltexte/2011/3068},
  URN =		{urn:nbn:de:0030-drops-30681},
  doi =		{10.4230/OASIcs.MEMICS.2010.77},
  annote =	{Keywords: JPEG2000, EBCOT, Context Modeling, GPU, GPGPU, parallel}
}

Keywords: JPEG2000, EBCOT, Context Modeling, GPU, GPGPU, parallel
Collection: Sixth Doctoral Workshop on Mathematical and Engineering Methods in Computer Science (MEMICS'10) -- Selected Papers
Issue Date: 2011
Date of publication: 11.03.2011

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI