Reducing power consumption of HEVC codec with lossless reference frame recompression

Dajiang Zhou; Li Guo; Jinjia Zhou; Satoshi Goto

doi:10.1109/ICIP.2014.7025425

Reducing power consumption of HEVC codec with lossless reference frame recompression

Dajiang Zhou, Li Guo, Jinjia Zhou, Satoshi Goto

Information, Production and Systems Research Center

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

12 Citations (Scopus)

Abstract

Motion estimation and motion compensation in HEVC and similar video codecs involve huge memory traffic in storing and loading reference frames. The resulting memory power composes a significant portion of system energy consumption. This paper presents a memory power reduction framework that losslessly compresses and decompresses reference frames on-the-fly. We first present the architecture that supports the random access of frame data compressed in variable ratios. The latest recompression algorithms and the corresponding VLSI implementation are also introduced. The framework has been implemented in several UHDTV ASIC and FPGA codecs. Experiments under HEVC common test conditions show an average memory traffic reduction of near 60%.

Original language	English
Title of host publication	2014 IEEE International Conference on Image Processing, ICIP 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2120-2124
Number of pages	5
ISBN (Print)	9781479957514
DOIs	https://doi.org/10.1109/ICIP.2014.7025425
Publication status	Published - 2014 Jan 28

Keywords

AVC
embedded compression
frame recompression
H.264
H.265
HEVC
lossless
power consumption
video decoder
video encoder

ASJC Scopus subject areas

Computer Vision and Pattern Recognition

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ICIP.2014.7025425

Cite this

Reducing power consumption of HEVC codec with lossless reference frame recompression. / Zhou, Dajiang; Guo, Li; Zhou, Jinjia et al.
2014 IEEE International Conference on Image Processing, ICIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 2120-2124 7025425.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

@inproceedings{8345d043a6a14597858fc4356d5f1b46,

title = "Reducing power consumption of HEVC codec with lossless reference frame recompression",

abstract = "Motion estimation and motion compensation in HEVC and similar video codecs involve huge memory traffic in storing and loading reference frames. The resulting memory power composes a significant portion of system energy consumption. This paper presents a memory power reduction framework that losslessly compresses and decompresses reference frames on-the-fly. We first present the architecture that supports the random access of frame data compressed in variable ratios. The latest recompression algorithms and the corresponding VLSI implementation are also introduced. The framework has been implemented in several UHDTV ASIC and FPGA codecs. Experiments under HEVC common test conditions show an average memory traffic reduction of near 60%.",

keywords = "AVC, embedded compression, frame recompression, H.264, H.265, HEVC, lossless, power consumption, video decoder, video encoder",

author = "Dajiang Zhou and Li Guo and Jinjia Zhou and Satoshi Goto",

year = "2014",

month = jan,

day = "28",

doi = "10.1109/ICIP.2014.7025425",

language = "English",

isbn = "9781479957514",

pages = "2120--2124",

booktitle = "2014 IEEE International Conference on Image Processing, ICIP 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Reducing power consumption of HEVC codec with lossless reference frame recompression

AU - Zhou, Dajiang

AU - Guo, Li

AU - Zhou, Jinjia

AU - Goto, Satoshi

PY - 2014/1/28

Y1 - 2014/1/28

N2 - Motion estimation and motion compensation in HEVC and similar video codecs involve huge memory traffic in storing and loading reference frames. The resulting memory power composes a significant portion of system energy consumption. This paper presents a memory power reduction framework that losslessly compresses and decompresses reference frames on-the-fly. We first present the architecture that supports the random access of frame data compressed in variable ratios. The latest recompression algorithms and the corresponding VLSI implementation are also introduced. The framework has been implemented in several UHDTV ASIC and FPGA codecs. Experiments under HEVC common test conditions show an average memory traffic reduction of near 60%.

AB - Motion estimation and motion compensation in HEVC and similar video codecs involve huge memory traffic in storing and loading reference frames. The resulting memory power composes a significant portion of system energy consumption. This paper presents a memory power reduction framework that losslessly compresses and decompresses reference frames on-the-fly. We first present the architecture that supports the random access of frame data compressed in variable ratios. The latest recompression algorithms and the corresponding VLSI implementation are also introduced. The framework has been implemented in several UHDTV ASIC and FPGA codecs. Experiments under HEVC common test conditions show an average memory traffic reduction of near 60%.

KW - AVC

KW - embedded compression

KW - frame recompression

KW - H.264

KW - H.265

KW - HEVC

KW - lossless

KW - power consumption

KW - video decoder

KW - video encoder

UR - http://www.scopus.com/inward/record.url?scp=84949926484&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949926484&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2014.7025425

DO - 10.1109/ICIP.2014.7025425

M3 - Conference contribution

AN - SCOPUS:84949926484

SN - 9781479957514

SP - 2120

EP - 2124

BT - 2014 IEEE International Conference on Image Processing, ICIP 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Reducing power consumption of HEVC codec with lossless reference frame recompression

Abstract

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this