We consider the reconstruction of a 3D object from measurements of liquid displacement, exploiting Archimedes law. These measurements are obtained by draining fluid from a container while the object is placed inside in various angles. The reconstruction is obtained by fitting simulations to the measurements, which leads to an inverse problem. To impose proper regularization, we use a parametric level set approach. We show that this way the reconstruction is obtained more accurately and efficiently than with other common approaches.

This presentation is part of Contributed Presentation “CP10 - Contributed session 10”

Authors:

Eran Treister (Ben Gurion University of the Negev )

Andrei Sharf (Ben Gurion University of the Negev)

Keywords:

computed tomography, computer graphics, inverse problems, nonlinear optimization