Stochastic Poisson Surface Reconstruction

We introduce a statistical extension of the classic Poisson Surface Reconstruction algorithm for recovering shapes from 3D point clouds. Instead outputting an implicit function, we represent reconstructed shape as modified Gaussian Process, which allows us to conduct queries (e.g., likelihood in space being on surface or inside solid). show that this perspective: improves PSR's integration into online scanning process, broadens its application realm, and opens door other lines research such applying task-specific priors.