Share Email Print

Optical Engineering • Open Access

Energy flow: image correspondence approximation for motion analysis
Author(s): Liangliang Wang; Ruifeng Li; Yajun Fang

Paper Abstract

We propose a correspondence approximation approach between temporally adjacent frames for motion analysis. First, energy map is established to represent image spatial features on multiple scales using Gaussian convolution. On this basis, energy flow at each layer is estimated using Gauss–Seidel iteration according to the energy invariance constraint. More specifically, at the core of energy invariance constraint is “energy conservation law” assuming that the spatial energy distribution of an image does not change significantly with time. Finally, energy flow field at different layers is reconstructed by considering different smoothness degrees. Due to the multiresolution origin and energy-based implementation, our algorithm is able to quickly address correspondence searching issues in spite of background noise or illumination variation. We apply our correspondence approximation method to motion analysis, and experimental results demonstrate its applicability.

Paper Details

Date Published: 29 April 2016
PDF: 8 pages
Opt. Eng. 55(4) 043109 doi: 10.1117/1.OE.55.4.043109
Published in: Optical Engineering Volume 55, Issue 4
Show Author Affiliations
Liangliang Wang, Harbin Institute of Technology (China)
Ruifeng Li, Harbin Institute of Technology (China)
Yajun Fang, Massachusetts Institute of Technology (United States)

© SPIE. Terms of Use
Back to Top