We present a pipeline for parametric wireframe extraction from denselysampled point clouds . Our approach processes a scalar distance field that represents proximity to the nearest sharp feature curve . In intermediatestages, it detects corners, constructs curve segmentation, and builds atopological graph fitted to wireframe . As an output, we produce parametricspline curves that can be edited and sampled arbitrarily . We evaluate ourmethod on 50 complex 3D shapes and compare it to the novel deep learning-basedtechnique .

Author(s) : Albert Matveev, Alexey Artemov, Denis Zorin, Evgeny Burnaev

Links : PDF - Abstract

Code :

Keywords : wireframe - distance - parametric - extraction - curve -

Leave a Reply

Your email address will not be published. Required fields are marked *