Building 3-D Human Face Models from Two Photographs
Journal of VLSI Signal Processing Systems - Special issue on multimedia signal processing
Hi-index | 0.00 |
We present a semiautomatic image-based modeling system to make 3-D models from photographs. The human aoperator intervenes only for simple and straightforward tasks. The design of the system is movtivated by potential applications where the geometric shapes fo the objects are complex, and the requirement on the appearance of the models is high. Easy to use is another consideration. We use binocular stereo as the basic model recovery tool, and propose a new volumetric matching algorithm that fully explores the spatial coherence between the two iamges. Inaccuate mathces are mostly clustered around the border area between the foreground and the background, thus easy to be manually removed. To make the system more practical, we insist on using uncalibrated cameras. Experimental resuls on different types of real objects are presented.