The longitudinal motion stereo problem: a CNN approach
S. Tarahlio, Andrea Zanela, A. Pellecchia · 2002
A cellular neural net (CNN) based approach to the resolution of the longitudinal motion stereo vision problem is presented. Because of the geometry of the system, through the use of a reference transform in the image sequence, the correlation of pixels between image frames can be performed with a static stereo vision algorithm, the Stereo-CNN. Results on real images are presented.