Optical flow from color image sequences

Robert J. Andrews (2004). Optical flow from color image sequences Master's Thesis, School of Engineering, The University of Queensland.

       
Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
THE17830.pdf Full text application/pdf 5.45MB 3
Author Robert J. Andrews
Thesis Title Optical flow from color image sequences
School, Centre or Institute School of Engineering
Institution The University of Queensland
Publication date 2004
Thesis type Master's Thesis
Supervisor Brian Lovell
Kurt Kubik
Total pages 99
Collection year 2004
Language eng
Subjects L
280208 Computer Vision
700199 Computer software and services not elsewhere classified
Formatted abstract

Recently proposed colour methods for optical flow recovery axe compared with traditional, gray-scale methods. New methods are presented which extend gray-scale methods into colour. Colour methods are shown to perform acceptably when compared to their gray-scale counterparts. 

During this research an extensible application was created to enable efficient comparison of various optical flow techniques by researchers and students. It features custom image sequences, synthetic ground-truth creation, custom pre-filtering, differentiation, optical flow methods (both gray-scale and colour) and flow analysis through image reconstruction, density/sparsity analysis, ground-truth analysis and flow visualisation. 

Keyword Optical character recognition devices

Document type: Thesis
Collection: UQ Theses (non-RHD) - UQ staff and students only
 
Citation counts: Google Scholar Search Google Scholar
Created: Fri, 24 Aug 2007, 18:25:37 EST