PLL Powered, Real-Time Visual Motion Tracking

We present a novel approach that handles the motion estimation/tracking problem in a phase-frequency tracking framework. Motion estimation in the spectral domain has been extensively studied. However, the real-time time-varying velocities tracking is still a challenging task. Our approach is based o...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on circuits and systems for video technology 2010-05, Vol.20 (5), p.632-646
Main Authors: Alexiadis, Dimitrios S, Sergiadis, George D
Format: Article
Language:English
Subjects:
Applied sciences
Artificial intelligence
Band pass filters
Cameras
Circuit properties
Circuits
Circuits of signal characteristics conditioning (including delay circuits)
Computer science
control theory
systems
Computer vision
Electric, optical and optoelectronic circuits
Electronic circuits
Electronics
Exact sciences and technology
Fourier transform
Hardware
Image processing
Image sequences
Information, signal and communications theory
Miscellaneous
Motion estimation
motion tracking
Navigation
Pattern recognition. Digital image processing. Computational geometry
Phase locked loops
Real time
Robot vision systems
Robustness
Signal processing
Spectra
Target tracking
Tasks
Telecommunications and information theory
Tracking
Visual
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a novel approach that handles the motion estimation/tracking problem in a phase-frequency tracking framework. Motion estimation in the spectral domain has been extensively studied. However, the real-time time-varying velocities tracking is still a challenging task. Our approach is based on a proposed phase-locked loop (PLL)-based system, which is hardware implementable and can be used for true real-time motion estimation of objects in image sequences. We describe all the necessary details of the PLL-based system which can track a single or multiple time-varying motions. We highlight the advantages and limitations of our methodology and we present promising simulation results on both synthetic and natural image sequences.
ISSN:1051-8215
1558-2205
DOI:10.1109/TCSVT.2010.2041832