Difference of Normals as a Multi-scale Operator in Unorganized Point Clouds

• Published in: 2012 Second International Conference on 3D Imaging, Modeling, Processing, Visualization & Transmission pp. 501 - 508
• Main Authors: Ioannou, Y., Taati, B., Harrap, R., Greenspan, M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-10-2012
• Subjects: 3D edges; computer vision; filtering; Image edge detection; Image segmentation; Kernel; KITTI; Laser radar; lidar; multi-scale; Noise; Object recognition; point cloud; segmentation; self-driving car; unorganized; unorganized point clouds; Vectors
• ISBN: 1467344702; 9781467344708
• ISSN: 1550-6185
• DOI: 10.1109/3DIMPVT.2012.12

A novel multi-scale operator for unorganized 3D point clouds is introduced. The Difference of Normals (DoN) provides a computationally efficient, multi-scale approach to processing large unorganized 3D point clouds. The application of DoN in the multi-scale filtering of two different real-world outdoor urban LIDAR scene datasets is quantitatively and qualitatively demonstrated. In both datasets the DoN operator is shown to segment large 3D point clouds into scale-salient clusters, such as cars, people, and lamp posts towards applications in semi-automatic annotation, and as a pre-processing step in automatic object recognition. The application of the operator to segmentation is evaluated on a large public dataset of outdoor LIDAR scenes with ground truth annotations.