Hierarchical motion estimation with smoothness constraints and postprocessing

Kan Xie; Luc Van Eycken; Andre J. Oosterlinck

doi:10.1117/1.600885

1 January 1996 Hierarchical motion estimation with smoothness constraints and postprocessing

Kan Xie, Luc Van Eycken, Andre J. Oosterlinck

Optical Engineering, Vol. 35, Issue 1, (January 1996). https://doi.org/10.1117/1.600885

How to acquire accurate and reliable motion parameters from an image sequence is a knotty problem for many applications in image processing, image recognition, and video coding, especially when scenes involve moving objects with various shapes and sizes as well as very fast and complicated motion. In this paper, an improved pel-based motion estimation (ME) algorithm with smoothness constraints is presented, which is based on the investigation and the comparison of different existing pel-based ME (or optical flow) algorithms. Then, in order to cope with various moving objects and their complex motion, a hierarchical ME algorithm with smoothness constraints and postprocessing is proposed. The experimental results show that the motion parameters obtained by the hierarchical ME algorithm are quite creditable and seem to be close to the real physical motion fields if the luminance intensity changes are due to the motion of objects. The hierarchical ME algorithm still provides approximate and smooth vector fields even for scenes that involve some motion-irrelevant intensity changes or blurring caused by violent motion.

Citation Download Citation

Kan Xie, Luc Van Eycken, and Andre J. Oosterlinck "Hierarchical motion estimation with smoothness constraints and postprocessing," Optical Engineering 35(1), (1 January 1996). https://doi.org/10.1117/1.600885

Published: 1 January 1996

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