Automatic Segmentation of Tree Structure From Point Cloud Data

In this project we developed a data-driven algorithm for segmenting the 3D point cloud of a tree into two main components, branches and leaves, by using geometric features computed directly on the point cloud.

A journal letter on this was published in the IEEE Robotics and Automation Letters (RAL), 2018 and presented at IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018.

(a) the input point cloud, (b) its extracted meso-skeleton and (c) the segmentation result of the proposed algorithm.