|project:||A Bayesian Approach for fitting Geometric Models in 3D Point Clouds|
I am a graduate student at the Bundeswehr University Munich.
PCL provides a great frame work for fitting geometric primitives e.g Planes, Cylinders, e.t.c. Most of these algorithms are based on the RANSAC paradigm. In this work however, we aim at solving the model fitting tasks using a Bayesian approach. Amongst others, this approach has the advantage of being very flexible and generic i.e. both parametric and non-parametric (e.g. curves, surfaces using splines) models can be fitted.
This section contains some of the “Bits and Bytes” from the discussions I’ve had with my mentor Zoltan