In this paper an efficient method for a rigid object shape recovery from a sequence of stereo images is presented. Unlike traditional multiview stereo approach our method reconstructs partial shape of the object from each pair of stereo images separately using stereovision technique. Then partial shapes, represented as point clouds, are fused together to recover full object geometry. Movement and/or rotation of the object in front of the stereo vision system doesn't need to be controlled. © 2011 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Komorowski, J. (2011). 3D object reconstruction from stereo images sequence. Advances in Intelligent and Soft Computing, 102, 199–207. https://doi.org/10.1007/978-3-642-23154-4_23
Mendeley helps you to discover research relevant for your work.