Title :
Inaccurate phase-based disparities removal
Author :
Ouali, M. ; Ziou, D. ; Laurgeau, C.
Author_Institution :
Dept. of Math. & Comput. Sci., Sherbrooke Univ., Que., Canada
Abstract :
Disparity estimation for 3D-structure reconstruction has become a classical topic in computer vision. Most of the recent work has been devoted to the phase-based techniques developed for their better theoretical grounding. One issue involved in the phase-based disparity estimation is the behavior of the phase which can produce some distorted results. Through the analysis of the responses we locate the inaccurate estimates and identify constraints such that they can be avoided. In this paper, we propose a constraint based on the magnitude of the responses to remove inaccurate disparity estimates and we give a technical proof to justify the constraint on the instantaneous frequency introduced before. The constraints used help to improve the accuracy of the measurements
Keywords :
computer vision; image matching; image reconstruction; 3D-structure reconstruction; computer vision; constraints; disparity estimation; inaccurate estimates; instantaneous frequency; phase-based disparities removal; Charge coupled devices; Computer science; Computer vision; Fourier transforms; Frequency estimation; Gabor filters; Image reconstruction; Layout; Mathematics; Phase distortion;
Conference_Titel :
Electronics, Circuits and Systems, 1999. Proceedings of ICECS '99. The 6th IEEE International Conference on
Conference_Location :
Pafos
Print_ISBN :
0-7803-5682-9
DOI :
10.1109/ICECS.1999.813395