Heterogeneous multiprocessor architecure for video coding applications

Publikation: Beitrag in FachzeitschriftKonferenzaufsatz in FachzeitschriftForschungPeer-Review

Autoren

  • Richard Hoffer
  • Winfried Gehrke
  • Peter Pirsch
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)417-424
Seitenumfang8
FachzeitschriftProceedings of SPIE - The International Society for Optical Engineering
Jahrgang1977
PublikationsstatusVeröffentlicht - 29 Okt. 1993
VeranstaltungVideo Communications and PACS for Medical Applications 1993 - Berlin, Deutschland
Dauer: 4 Apr. 19939 Apr. 1993

Abstract

A multiprocessor architecture for compact realizations of video coding applications is presented. The actual standards for video coding e.g. H.261 and MPEG are based on a hybrid coding scheme, which allows parallelization at both data level and task level. The parallelization at data level is performed by distribution of image data among the processors. Each processor works on locally stored image segments. The parallelization at task level is realized inside the processors by functional modules which are adapted to classes of algorithms. The functionality of the modules and the number of their data paths is determined by applying efficiency calculations resulting in a module for motion estimation and a blocklevel coprocessor for transform and quantization. The controlling and synchronization is accomplished by a programmable module. A hierarchical controlling concept reduces the on-chip control overhead. A chip size of 70 mm2 is estimated for one processor, when using 0.6 μm CMOS technology. With an operating frequency of 65 MHz one chip will perform the computations for a full CIF H.261 codec with 30 Hz framerate and motion estimation based on +/-15 pel full search blockmatching algorithm.

ASJC Scopus Sachgebiete

Zitieren

Heterogeneous multiprocessor architecure for video coding applications. / Hoffer, Richard; Gehrke, Winfried; Pirsch, Peter.
in: Proceedings of SPIE - The International Society for Optical Engineering, Jahrgang 1977, 29.10.1993, S. 417-424.

Publikation: Beitrag in FachzeitschriftKonferenzaufsatz in FachzeitschriftForschungPeer-Review

Hoffer, R, Gehrke, W & Pirsch, P 1993, 'Heterogeneous multiprocessor architecure for video coding applications', Proceedings of SPIE - The International Society for Optical Engineering, Jg. 1977, S. 417-424. https://doi.org/10.1117/12.160486
Hoffer, R., Gehrke, W., & Pirsch, P. (1993). Heterogeneous multiprocessor architecure for video coding applications. Proceedings of SPIE - The International Society for Optical Engineering, 1977, 417-424. https://doi.org/10.1117/12.160486
Hoffer R, Gehrke W, Pirsch P. Heterogeneous multiprocessor architecure for video coding applications. Proceedings of SPIE - The International Society for Optical Engineering. 1993 Okt 29;1977:417-424. doi: 10.1117/12.160486
Hoffer, Richard ; Gehrke, Winfried ; Pirsch, Peter. / Heterogeneous multiprocessor architecure for video coding applications. in: Proceedings of SPIE - The International Society for Optical Engineering. 1993 ; Jahrgang 1977. S. 417-424.
Download
@article{b4df4ec2146e417d992a642d8c801f5d,
title = "Heterogeneous multiprocessor architecure for video coding applications",
abstract = "A multiprocessor architecture for compact realizations of video coding applications is presented. The actual standards for video coding e.g. H.261 and MPEG are based on a hybrid coding scheme, which allows parallelization at both data level and task level. The parallelization at data level is performed by distribution of image data among the processors. Each processor works on locally stored image segments. The parallelization at task level is realized inside the processors by functional modules which are adapted to classes of algorithms. The functionality of the modules and the number of their data paths is determined by applying efficiency calculations resulting in a module for motion estimation and a blocklevel coprocessor for transform and quantization. The controlling and synchronization is accomplished by a programmable module. A hierarchical controlling concept reduces the on-chip control overhead. A chip size of 70 mm2 is estimated for one processor, when using 0.6 μm CMOS technology. With an operating frequency of 65 MHz one chip will perform the computations for a full CIF H.261 codec with 30 Hz framerate and motion estimation based on +/-15 pel full search blockmatching algorithm.",
author = "Richard Hoffer and Winfried Gehrke and Peter Pirsch",
note = "Funding Information: This work is supported by the Corporate Research and Development Siemens AG, Munich. ; Video Communications and PACS for Medical Applications 1993 ; Conference date: 04-04-1993 Through 09-04-1993",
year = "1993",
month = oct,
day = "29",
doi = "10.1117/12.160486",
language = "English",
volume = "1977",
pages = "417--424",

}

Download

TY - JOUR

T1 - Heterogeneous multiprocessor architecure for video coding applications

AU - Hoffer, Richard

AU - Gehrke, Winfried

AU - Pirsch, Peter

N1 - Funding Information: This work is supported by the Corporate Research and Development Siemens AG, Munich.

PY - 1993/10/29

Y1 - 1993/10/29

N2 - A multiprocessor architecture for compact realizations of video coding applications is presented. The actual standards for video coding e.g. H.261 and MPEG are based on a hybrid coding scheme, which allows parallelization at both data level and task level. The parallelization at data level is performed by distribution of image data among the processors. Each processor works on locally stored image segments. The parallelization at task level is realized inside the processors by functional modules which are adapted to classes of algorithms. The functionality of the modules and the number of their data paths is determined by applying efficiency calculations resulting in a module for motion estimation and a blocklevel coprocessor for transform and quantization. The controlling and synchronization is accomplished by a programmable module. A hierarchical controlling concept reduces the on-chip control overhead. A chip size of 70 mm2 is estimated for one processor, when using 0.6 μm CMOS technology. With an operating frequency of 65 MHz one chip will perform the computations for a full CIF H.261 codec with 30 Hz framerate and motion estimation based on +/-15 pel full search blockmatching algorithm.

AB - A multiprocessor architecture for compact realizations of video coding applications is presented. The actual standards for video coding e.g. H.261 and MPEG are based on a hybrid coding scheme, which allows parallelization at both data level and task level. The parallelization at data level is performed by distribution of image data among the processors. Each processor works on locally stored image segments. The parallelization at task level is realized inside the processors by functional modules which are adapted to classes of algorithms. The functionality of the modules and the number of their data paths is determined by applying efficiency calculations resulting in a module for motion estimation and a blocklevel coprocessor for transform and quantization. The controlling and synchronization is accomplished by a programmable module. A hierarchical controlling concept reduces the on-chip control overhead. A chip size of 70 mm2 is estimated for one processor, when using 0.6 μm CMOS technology. With an operating frequency of 65 MHz one chip will perform the computations for a full CIF H.261 codec with 30 Hz framerate and motion estimation based on +/-15 pel full search blockmatching algorithm.

UR - http://www.scopus.com/inward/record.url?scp=0346050893&partnerID=8YFLogxK

U2 - 10.1117/12.160486

DO - 10.1117/12.160486

M3 - Conference article

AN - SCOPUS:0346050893

VL - 1977

SP - 417

EP - 424

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

SN - 0277-786X

T2 - Video Communications and PACS for Medical Applications 1993

Y2 - 4 April 1993 through 9 April 1993

ER -