The mean field theory for image motion estimation

Authors:
J. Zhang;J. Hanauer
Affiliations:
Electrical Engineering and Computer Science Department, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin;Electrical Engineering and Computer Science Department, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin
Venue:
ICASSP'93 Proceedings of the 1993 IEEE international conference on Acoustics, speech, and signal processing: image and multidimensional signal processing - Volume V
Year:
1993

Citing 4
Cited 0

Energy minimization approach to motion estimation

Signal Processing - Special issue on multidimensional signal processing
Robot Vision

Robot Vision
Parallel and Deterministic Algorithms for MRFs: Surface Reconstruction and Integration

Parallel and Deterministic Algorithms for MRFs: Surface Reconstruction and Integration
Comparison of stochastic and deterministic solution methods in Bayesian estimation of 2D motion

Image and Vision Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recently, the mean field theory (MFT) has been shown to be agood alternative to simulated annealing (SA) in optimization problems related to Markov random fields (MRF); it provides comparable performance while converging much faster. In this work, we show how the MFT can be applied to MRF model-based motion estimation. Specifically, the motion is characterized by a coupled MRF including a displacement field (motion continuity), a line field (motion discontinuity), and a segmentation field (identifying uncovered areas). These fields are estimated by using MFT. The efficacy of this approach is demonstrated on synthetic and real-world images.