A fast block motion estimation algorithm using Gray Code Kernels

Yair Moshe; Hagit Hel-Or

doi:10.1109/ISSPIT.2006.270794

A fast block motion estimation algorithm using Gray Code Kernels

Yair Moshe, Hagit Hel-Or

Department of Computer Science

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

Abstract

Motion estimation plays an important role in modern video coders. In such coders, motion is estimated using a block matching algorithm that estimates the amount of motion on a block-by-black basis. A full search technique for finding the best matching blocks delivers good accuracy but is usually not practical due to its high computational complexity. In this paper, a novel fast block-based motion estimation algorithm is proposed. This algorithm uses an efficient projection framework which bounds the distance between a template block and candidate blocks. Fast projection is performed with a family of highly efficient filter kernels - the Gray Code Kernels - using only 2 operations per pixel for each filter kernel. The projection framework is combined with a rejection scheme which allows rapid rejection of candidate blocks that are distant from the template block. Experiments show that the proposed algorithm significantly outperforms popular fast motion estimation algorithms, such as three-step search and diamond search. In addition, the tradeoff between computational complexity and quality of results could be easily controlled in the proposed algorithm, thus it enables adaptivity to image content.

Original language	English
Title of host publication	Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	185-190
Number of pages	6
ISBN (Print)	0780397541, 9780780397545
DOIs	https://doi.org/10.1109/ISSPIT.2006.270794
State	Published - 2006
Event	6th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006 - Vancouver, BC, Canada Duration: 27 Aug 2006 → 30 Aug 2006

Publication series

Name	Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT

Conference

Conference	6th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006
Country/Territory	Canada
City	Vancouver, BC
Period	27/08/06 → 30/08/06

Keywords

Block matching
Fast motion estimation
Gray Code Kernels (GCK)
Video coding
Video compression

ASJC Scopus subject areas

Signal Processing
Software
Computer Networks and Communications

Access to Document

10.1109/ISSPIT.2006.270794

Cite this

Moshe, Y., & Hel-Or, H. (2006). A fast block motion estimation algorithm using Gray Code Kernels. In Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006 (pp. 185-190). Article 4042236 (Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISSPIT.2006.270794

@inproceedings{cdb417254adb4e9f9d63b39f81d2da98,

title = "A fast block motion estimation algorithm using Gray Code Kernels",

abstract = "Motion estimation plays an important role in modern video coders. In such coders, motion is estimated using a block matching algorithm that estimates the amount of motion on a block-by-black basis. A full search technique for finding the best matching blocks delivers good accuracy but is usually not practical due to its high computational complexity. In this paper, a novel fast block-based motion estimation algorithm is proposed. This algorithm uses an efficient projection framework which bounds the distance between a template block and candidate blocks. Fast projection is performed with a family of highly efficient filter kernels - the Gray Code Kernels - using only 2 operations per pixel for each filter kernel. The projection framework is combined with a rejection scheme which allows rapid rejection of candidate blocks that are distant from the template block. Experiments show that the proposed algorithm significantly outperforms popular fast motion estimation algorithms, such as three-step search and diamond search. In addition, the tradeoff between computational complexity and quality of results could be easily controlled in the proposed algorithm, thus it enables adaptivity to image content.",

keywords = "Block matching, Fast motion estimation, Gray Code Kernels (GCK), Video coding, Video compression",

author = "Yair Moshe and Hagit Hel-Or",

year = "2006",

doi = "10.1109/ISSPIT.2006.270794",

language = "English",

isbn = "0780397541",

series = "Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT",

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

pages = "185--190",

booktitle = "Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006",

address = "United States",

note = "6th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006 ; Conference date: 27-08-2006 Through 30-08-2006",

}

Moshe, Y & Hel-Or, H 2006, A fast block motion estimation algorithm using Gray Code Kernels. in Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006., 4042236, Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT, Institute of Electrical and Electronics Engineers Inc., pp. 185-190, 6th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006, Vancouver, BC, Canada, 27/08/06. https://doi.org/10.1109/ISSPIT.2006.270794

A fast block motion estimation algorithm using Gray Code Kernels. / Moshe, Yair; Hel-Or, Hagit.
Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006. Institute of Electrical and Electronics Engineers Inc., 2006. p. 185-190 4042236 (Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT).

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

TY - GEN

T1 - A fast block motion estimation algorithm using Gray Code Kernels

AU - Moshe, Yair

AU - Hel-Or, Hagit

PY - 2006

Y1 - 2006

N2 - Motion estimation plays an important role in modern video coders. In such coders, motion is estimated using a block matching algorithm that estimates the amount of motion on a block-by-black basis. A full search technique for finding the best matching blocks delivers good accuracy but is usually not practical due to its high computational complexity. In this paper, a novel fast block-based motion estimation algorithm is proposed. This algorithm uses an efficient projection framework which bounds the distance between a template block and candidate blocks. Fast projection is performed with a family of highly efficient filter kernels - the Gray Code Kernels - using only 2 operations per pixel for each filter kernel. The projection framework is combined with a rejection scheme which allows rapid rejection of candidate blocks that are distant from the template block. Experiments show that the proposed algorithm significantly outperforms popular fast motion estimation algorithms, such as three-step search and diamond search. In addition, the tradeoff between computational complexity and quality of results could be easily controlled in the proposed algorithm, thus it enables adaptivity to image content.

AB - Motion estimation plays an important role in modern video coders. In such coders, motion is estimated using a block matching algorithm that estimates the amount of motion on a block-by-black basis. A full search technique for finding the best matching blocks delivers good accuracy but is usually not practical due to its high computational complexity. In this paper, a novel fast block-based motion estimation algorithm is proposed. This algorithm uses an efficient projection framework which bounds the distance between a template block and candidate blocks. Fast projection is performed with a family of highly efficient filter kernels - the Gray Code Kernels - using only 2 operations per pixel for each filter kernel. The projection framework is combined with a rejection scheme which allows rapid rejection of candidate blocks that are distant from the template block. Experiments show that the proposed algorithm significantly outperforms popular fast motion estimation algorithms, such as three-step search and diamond search. In addition, the tradeoff between computational complexity and quality of results could be easily controlled in the proposed algorithm, thus it enables adaptivity to image content.

KW - Block matching

KW - Fast motion estimation

KW - Gray Code Kernels (GCK)

KW - Video coding

KW - Video compression

UR - http://www.scopus.com/inward/record.url?scp=44449143541&partnerID=8YFLogxK

U2 - 10.1109/ISSPIT.2006.270794

DO - 10.1109/ISSPIT.2006.270794

M3 - Conference contribution

AN - SCOPUS:44449143541

SN - 0780397541

SN - 9780780397545

T3 - Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT

SP - 185

EP - 190

BT - Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 6th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006

Y2 - 27 August 2006 through 30 August 2006

ER -

Moshe Y, Hel-Or H. A fast block motion estimation algorithm using Gray Code Kernels. In Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2006. Institute of Electrical and Electronics Engineers Inc. 2006. p. 185-190. 4042236. (Sixth IEEE International Symposium on Signal Processing and Information Technology, ISSPIT). doi: 10.1109/ISSPIT.2006.270794

A fast block motion estimation algorithm using Gray Code Kernels

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this