Proximity Computations between Noisy Point Clouds using Robust Classification

We present a new approach to perform robust proximity queries between noisy point cloud data. Our approach takes into account the uncertainty that arises due to discretization error and noise, and formulates contact computation as a twoclass classification problem. We use appropriate techniques from machine learning to compute the collision probability for each point in the input data and accelerate the computation using stochastic traversal of bounding volume hierarchies. We highlight the performance of our algorithm on point clouds captured using laser range finders, active stereo cameras and synthetic datasets.