Adaptive Reconstruction of Surfaces and Scalar Fields from Dense Scattered Trivariate Data •

In this paper, we propose an approach to model a set oftrivariate dense and scattered dalaon an unknown manifold by the zero contour of the a l piecewise tri-quadratic or tri-cubic polynomial. The 0 1function approximates the data over an octrec·like mesh in an adaptive fashion. At the same time, and applying the sllme techniques. a scalar field (scalar values associated with the sampled surface points) is also modeled as a C 1 piecewise polynomial function defined on the reconstructed surface. The reconstructed surface and field can be used to visualize and interact with the data in a variety of ways.