Adaptive correlation filter-based video stabilization without accumulative global motion estimation

Eunjin Koh; Chanyong Lee; Dong Gil Jeong

doi:10.1117/1.OE.53.12.123104

8 December 2014 Adaptive correlation filter-based video stabilization without accumulative global motion estimation

Eunjin Koh, Chanyong Lee, Dong Gil Jeong

Author Affiliations +

Optical Engineering, Vol. 53, Issue 12, 123104 (December 2014). https://doi.org/10.1117/1.OE.53.12.123104

Abstract

We present a digital video stabilization approach that provides both robustness and efficiency for practical applications. In this approach, we adopt a stabilization model that maintains spatio-temporal information of past input frames efficiently and can track original stabilization position. Because of the stabilization model, the proposed method does not need accumulative global motion estimation and can recover the original position even if there is a failure in interframe motion estimation. It can also intelligently overcome the situation of damaged or interrupted video sequences. Moreover, because it is simple and suitable to parallel scheme, we implement it on a commercial field programmable gate array and a graphics processing unit board with compute unified device architecture in a breeze. Experimental results show that the proposed approach is both fast and robust.

Citation Download Citation

Eunjin Koh, Chanyong Lee, and Dong Gil Jeong "Adaptive correlation filter-based video stabilization without accumulative global motion estimation," Optical Engineering 53(12), 123104 (8 December 2014). https://doi.org/10.1117/1.OE.53.12.123104

Published: 8 December 2014

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available