Improved optical flow motion estimation for digital image stabilization

Lijun Lai; Zhiyong Xu; Xuyao Zhang

doi:10.1117/12.2216083

5 November 2015 Improved optical flow motion estimation for digital image stabilization

Lijun Lai, Zhiyong Xu, Xuyao Zhang

Proceedings Volume 9795, Selected Papers of the Photoelectronic Technology Committee Conferences held June–July 2015; 97951Q (2015) https://doi.org/10.1117/12.2216083
Event: Selected Proceedings of the Photoelectronic Technology Committee Conferences held June-July 2015, 2015, Hefei, Suzhou, and Harbin, China

Abstract

Optical flow is the instantaneous motion vector at each pixel in the image frame at a time instant. The gradient-based approach for optical flow computation can’t work well when the video motion is too large. To alleviate such problem, we incorporate this algorithm into a pyramid multi-resolution coarse-to-fine search strategy. Using pyramid strategy to obtain multi-resolution images; Using iterative relationship from the highest level to the lowest level to obtain inter-frames’ affine parameters; Subsequence frames compensate back to the first frame to obtain stabilized sequence. The experiment results demonstrate that the promoted method has good performance in global motion estimation.

Citation Download Citation

Lijun Lai, Zhiyong Xu, and Xuyao Zhang "Improved optical flow motion estimation for digital image stabilization", Proc. SPIE 9795, Selected Papers of the Photoelectronic Technology Committee Conferences held June–July 2015, 97951Q (5 November 2015); https://doi.org/10.1117/12.2216083

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