Year: 2010
Type: conference-paper
Abstract: We describe a computationally fast and effective approach to 2D-3D conversion of an image pair for the threedimensional rendering on stereoscopic displays of scenes \nincluding a ground plane. The stereo disparities of all the \nother scene elements (background, foreground objects) are \ncomputed after statistical segmentation and geometric localization of the ground plane. Geometric estimation includes camera self-calibration from epipolar geometry, and \nan original algorithm for the recovery of 3D visual parameters from the properties of planar homologies. Experimental results on real images show that, notwithstanding \nthe simple “theatrical” model employed for the scene, the \ndisparity maps generated with our approach are accurate \nenough to provide users with a stunning 3D impression of \nthe displayed scene, and fast enough to be extended to video \nsequences.
Institutions: Kyoto University, University of Florence, University of Cambridge
Language: English
Topic: Advanced Vision and Imaging
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3159899904