Meshing Point Clouds using spherical parameterization
3D scanners produce sets of 3D data points, sampled from the surface of a 3D object. These points are frequently unorganized, and to use them in 3D applications requires computing a polygon (usually triangular) mesh which best approximates the sampled surface. This means associating a connectivity structure with the point set.
This project addresses the problem of meshing point clouds using spherical parameterization method presented in Gotsman's paper here http://www.cs.technion.ac.il/~gotsman/AmendedPubl/Matthias/meshing.pdf.
Our approach relies on the automatic finding of the correspondences between the landmarks on the input laser scan and a template face mesh. Once we find the set of correspondences, we define a morphing function and apply it to the template mesh. As a result, the template mesh is morphed and represents the same facial features captured from the laser scan. Our method is completely automatic and does not rely neither on:
- any prior anatomical knowledge about the position of the landmarks or on
- the morphology of the input face.
Usage
library used in the porject:
- Boost, Qt
- OpenMesh
- CGAL Lib
- Profiler
- Qhull
- Eigen 3.0
- visual studio
