DocumentCode
3139408
Title
Relative depth from motion using normal flow: an active and purposive solution
Author
Huang, Liuqing ; Aloimonos, Yiannis
Author_Institution
Maryland Univ., College Park, MD, USA
fYear
1991
fDate
7-9 Oct 1991
Firstpage
196
Lastpage
204
Abstract
The authors show how an active observer can compute the relative depth of (stationary or moving) objects in the field of view using only the spatiotemporal derivatives of the time varying image intensity function. The solution they propose is purposive in the sense that it solves only the relative depth from motion problem and cannot be used for other problems related to motion; active in the sense that the activity of the observer is essential for the solution of the problem. Results indicate that exact computation of retinal motion does not appear to be a necessary first step for some problems related to visual motion. In addition, optic flow, whose computation is an ill-posed problem, is related to the motion of the scene only under very restrictive assumptions. As a result, the use of optic flow in some quantitative motion analysis studies is questionable
Keywords
image recognition; image sequences; motion estimation; 3D information; normal flow; optic flow; relative depth; retinal motion; spatiotemporal derivatives; time varying image intensity function; visual motion; Automation; Computer vision; Educational institutions; Image motion analysis; Laboratories; Layout; Motion analysis; Nonlinear optics; Optical computing; Optical sensors;
fLanguage
English
Publisher
ieee
Conference_Titel
Visual Motion, 1991., Proceedings of the IEEE Workshop on
Conference_Location
Princeton, NJ
Print_ISBN
0-8186-2153-2
Type
conf
DOI
10.1109/WVM.1991.212807
Filename
212807
Link To Document