Lucas-Kanade without iterative warping

Alex Rav-Acha; Shmuel Peleg

doi:10.1109/ICIP.2006.312747

Lucas-Kanade without iterative warping

Alex Rav-Acha^*, Shmuel Peleg

^*Corresponding author for this work

The Rachel and Selim Benin School of Engineering and Computer Science

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

16 Scopus citations

Abstract

Many methods for motion computation and object tracking are based on the Lucas-Kanade (LK) framework [1]. We present a method which substantially speeds up the LK approach while preserving its accuracy. This acceleration is obtained by avoiding the iterative image warping, inherent to the LK framework. A three-fold speedup is observed on standard image alignment tasks. Our second contribution focuses on adopting a multi-frame approach in order to increase alignment accuracy and robustness. By utilizing the acceleration procedure, the complexity of this multi-frame alignment becomes comparable to that of the two-frame approach.

Original language	English
Title of host publication	2006 IEEE International Conference on Image Processing, ICIP 2006 - Proceedings
Pages	1097-1100
Number of pages	4
DOIs	https://doi.org/10.1109/ICIP.2006.312747
State	Published - 2006
Event	2006 IEEE International Conference on Image Processing, ICIP 2006 - Atlanta, GA, United States Duration: 8 Oct 2006 → 11 Oct 2006

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
ISSN (Print)	1522-4880

Conference

Conference	2006 IEEE International Conference on Image Processing, ICIP 2006
Country/Territory	United States
City	Atlanta, GA
Period	8/10/06 → 11/10/06

Keywords

Direct methods
Image alignment
Motion analysis
Video stabilization

Access to Document

10.1109/ICIP.2006.312747

Cite this

@inproceedings{f398d5380f534241b7c6139434939bf9,

title = "Lucas-Kanade without iterative warping",

abstract = "Many methods for motion computation and object tracking are based on the Lucas-Kanade (LK) framework [1]. We present a method which substantially speeds up the LK approach while preserving its accuracy. This acceleration is obtained by avoiding the iterative image warping, inherent to the LK framework. A three-fold speedup is observed on standard image alignment tasks. Our second contribution focuses on adopting a multi-frame approach in order to increase alignment accuracy and robustness. By utilizing the acceleration procedure, the complexity of this multi-frame alignment becomes comparable to that of the two-frame approach.",

keywords = "Direct methods, Image alignment, Motion analysis, Video stabilization",

author = "Alex Rav-Acha and Shmuel Peleg",

year = "2006",

doi = "10.1109/ICIP.2006.312747",

language = "אנגלית",

isbn = "1424404819",

series = "Proceedings - International Conference on Image Processing, ICIP",

pages = "1097--1100",

booktitle = "2006 IEEE International Conference on Image Processing, ICIP 2006 - Proceedings",

note = "2006 IEEE International Conference on Image Processing, ICIP 2006 ; Conference date: 08-10-2006 Through 11-10-2006",

}

Rav-Acha, A & Peleg, S 2006, Lucas-Kanade without iterative warping. in 2006 IEEE International Conference on Image Processing, ICIP 2006 - Proceedings., 4106725, Proceedings - International Conference on Image Processing, ICIP, pp. 1097-1100, 2006 IEEE International Conference on Image Processing, ICIP 2006, Atlanta, GA, United States, 8/10/06. https://doi.org/10.1109/ICIP.2006.312747

TY - GEN

T1 - Lucas-Kanade without iterative warping

AU - Rav-Acha, Alex

AU - Peleg, Shmuel

PY - 2006

Y1 - 2006

N2 - Many methods for motion computation and object tracking are based on the Lucas-Kanade (LK) framework [1]. We present a method which substantially speeds up the LK approach while preserving its accuracy. This acceleration is obtained by avoiding the iterative image warping, inherent to the LK framework. A three-fold speedup is observed on standard image alignment tasks. Our second contribution focuses on adopting a multi-frame approach in order to increase alignment accuracy and robustness. By utilizing the acceleration procedure, the complexity of this multi-frame alignment becomes comparable to that of the two-frame approach.

AB - Many methods for motion computation and object tracking are based on the Lucas-Kanade (LK) framework [1]. We present a method which substantially speeds up the LK approach while preserving its accuracy. This acceleration is obtained by avoiding the iterative image warping, inherent to the LK framework. A three-fold speedup is observed on standard image alignment tasks. Our second contribution focuses on adopting a multi-frame approach in order to increase alignment accuracy and robustness. By utilizing the acceleration procedure, the complexity of this multi-frame alignment becomes comparable to that of the two-frame approach.

KW - Direct methods

KW - Image alignment

KW - Motion analysis

KW - Video stabilization

UR - http://www.scopus.com/inward/record.url?scp=40949113943&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2006.312747

DO - 10.1109/ICIP.2006.312747

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:40949113943

SN - 1424404819

SN - 9781424404810

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 1097

EP - 1100

BT - 2006 IEEE International Conference on Image Processing, ICIP 2006 - Proceedings

T2 - 2006 IEEE International Conference on Image Processing, ICIP 2006

Y2 - 8 October 2006 through 11 October 2006

ER -

Lucas-Kanade without iterative warping

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this