Robotics: Science and Systems III

Context and Feature Sensitive Re-sampling from Discrete Surface Measurements

Dave Cole and Paul Newman

Abstract: This paper concerns context and feature-sensitive sampling of workspace surfaces by processing 3D point clouds. We pay particular attention to working with data from 3D laser range finders. We interpret the point cloud as the outcome of repetitive and non-uniform sampling of the surfaces in the workspace. The nature of this sampling may not be ideal for all applications, representations and downstream processing. For example it might be preferable to have a high point density around sharp edges or near marked changes in texture. Additionally such preferences might be dependent on the semantic classification of the surface in question. This paper addresses this issue and provides a framework which given a raw point cloud as input, produces a new point cloud by sampling from the underlying workspace surfaces. Moreover it does this in a manner which can be biased by local low-level geometric or appearance properties and higher level (semantic) classification of the surface. We are in no way prescriptive about what justifies a biasing in the sampling scheme --- this is left up to the user who may encapsulate what constitutes ``interesting'' into one or more ``policies'' which are used to modulate the default sampling behavior.



    AUTHOR    = {D. Cole and P. Newman},
    TITLE     = {Context and Feature Sensitive Re-sampling from Discrete Surface Measurements},
    BOOKTITLE = {Proceedings of Robotics: Science and Systems},
    YEAR      = {2007},
    ADDRESS   = {Atlanta, GA, USA},
    MONTH     = {June},
    DOI       = {10.15607/RSS.2007.III.013}