A fine-grain scalable and low memory cost variable block size motion estimation architecture for H.264/AVC

Zhenyu Liu, Yang Song, Takeshi Ikenaga, Satoshi Goto

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

One full search variable block size motion estimation (VBSME) architecture with integer pixel accuracy is proposed in this paper. This proposed architecture has following features: (1) Through widening data path from the search area memories, m processing element groups (PEG) could be scheduled to work in parallel and fully utilized, where m is a factor of sixteen. Each PEG has sixteen processing elements (PE) and just costs 8.5K gates. This feature provides users more flexibility to make tradeoff between the hardware cost and the performance. (2) Based on pipelining and multi-cycle data path techniques, this architecture can work at high clock frequency. (3) The memory partition number is greatly reduced. When sixteen PEGs are adopted, only two memory partitions are required for the search area data storage. Therefore, both the system hardware cost and power consumption can be saved. A 16-PEG design with 48 × 32 search range has been implemented with TSMC 0.18 μm CMOS technology. In typical work conditions, its maximum clock frequency is 261 MHz. Compared with the previous 2-D architecture [9], about 13.4 hardware cost and 5.7 power consumption can be saved.

Original languageEnglish
Pages (from-to)1928-1936
Number of pages9
JournalIEICE Transactions on Electronics
VolumeE89-C
Issue number12
DOIs
Publication statusPublished - 2006 Dec

Fingerprint

Motion estimation
Data storage equipment
Processing
Costs
Clocks
Electric power utilization
Hardware
Computer hardware
Polyethylene glycols
Computer systems
Pixels

Keywords

  • AVC
  • H.264
  • Variable block size motion estimation
  • VLSI architecture

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

A fine-grain scalable and low memory cost variable block size motion estimation architecture for H.264/AVC. / Liu, Zhenyu; Song, Yang; Ikenaga, Takeshi; Goto, Satoshi.

In: IEICE Transactions on Electronics, Vol. E89-C, No. 12, 12.2006, p. 1928-1936.

Research output: Contribution to journalArticle

@article{392fa9067af645e8aac949af1d2eba0b,
title = "A fine-grain scalable and low memory cost variable block size motion estimation architecture for H.264/AVC",
abstract = "One full search variable block size motion estimation (VBSME) architecture with integer pixel accuracy is proposed in this paper. This proposed architecture has following features: (1) Through widening data path from the search area memories, m processing element groups (PEG) could be scheduled to work in parallel and fully utilized, where m is a factor of sixteen. Each PEG has sixteen processing elements (PE) and just costs 8.5K gates. This feature provides users more flexibility to make tradeoff between the hardware cost and the performance. (2) Based on pipelining and multi-cycle data path techniques, this architecture can work at high clock frequency. (3) The memory partition number is greatly reduced. When sixteen PEGs are adopted, only two memory partitions are required for the search area data storage. Therefore, both the system hardware cost and power consumption can be saved. A 16-PEG design with 48 × 32 search range has been implemented with TSMC 0.18 μm CMOS technology. In typical work conditions, its maximum clock frequency is 261 MHz. Compared with the previous 2-D architecture [9], about 13.4 hardware cost and 5.7 power consumption can be saved.",
keywords = "AVC, H.264, Variable block size motion estimation, VLSI architecture",
author = "Zhenyu Liu and Yang Song and Takeshi Ikenaga and Satoshi Goto",
year = "2006",
month = "12",
doi = "10.1093/ietele/e89-c.12.1928",
language = "English",
volume = "E89-C",
pages = "1928--1936",
journal = "IEICE Transactions on Electronics",
issn = "0916-8524",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "12",

}

TY - JOUR

T1 - A fine-grain scalable and low memory cost variable block size motion estimation architecture for H.264/AVC

AU - Liu, Zhenyu

AU - Song, Yang

AU - Ikenaga, Takeshi

AU - Goto, Satoshi

PY - 2006/12

Y1 - 2006/12

N2 - One full search variable block size motion estimation (VBSME) architecture with integer pixel accuracy is proposed in this paper. This proposed architecture has following features: (1) Through widening data path from the search area memories, m processing element groups (PEG) could be scheduled to work in parallel and fully utilized, where m is a factor of sixteen. Each PEG has sixteen processing elements (PE) and just costs 8.5K gates. This feature provides users more flexibility to make tradeoff between the hardware cost and the performance. (2) Based on pipelining and multi-cycle data path techniques, this architecture can work at high clock frequency. (3) The memory partition number is greatly reduced. When sixteen PEGs are adopted, only two memory partitions are required for the search area data storage. Therefore, both the system hardware cost and power consumption can be saved. A 16-PEG design with 48 × 32 search range has been implemented with TSMC 0.18 μm CMOS technology. In typical work conditions, its maximum clock frequency is 261 MHz. Compared with the previous 2-D architecture [9], about 13.4 hardware cost and 5.7 power consumption can be saved.

AB - One full search variable block size motion estimation (VBSME) architecture with integer pixel accuracy is proposed in this paper. This proposed architecture has following features: (1) Through widening data path from the search area memories, m processing element groups (PEG) could be scheduled to work in parallel and fully utilized, where m is a factor of sixteen. Each PEG has sixteen processing elements (PE) and just costs 8.5K gates. This feature provides users more flexibility to make tradeoff between the hardware cost and the performance. (2) Based on pipelining and multi-cycle data path techniques, this architecture can work at high clock frequency. (3) The memory partition number is greatly reduced. When sixteen PEGs are adopted, only two memory partitions are required for the search area data storage. Therefore, both the system hardware cost and power consumption can be saved. A 16-PEG design with 48 × 32 search range has been implemented with TSMC 0.18 μm CMOS technology. In typical work conditions, its maximum clock frequency is 261 MHz. Compared with the previous 2-D architecture [9], about 13.4 hardware cost and 5.7 power consumption can be saved.

KW - AVC

KW - H.264

KW - Variable block size motion estimation

KW - VLSI architecture

UR - http://www.scopus.com/inward/record.url?scp=33845594880&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845594880&partnerID=8YFLogxK

U2 - 10.1093/ietele/e89-c.12.1928

DO - 10.1093/ietele/e89-c.12.1928

M3 - Article

VL - E89-C

SP - 1928

EP - 1936

JO - IEICE Transactions on Electronics

JF - IEICE Transactions on Electronics

SN - 0916-8524

IS - 12

ER -