NormNet: Point-wise normal estimation network for three-dimensional point cloud data. (4th July 2019)

• Record Type: Journal Article
• Authors: Hyeon, Janghun; Lee, Weonsuk; Kim, Joo Hyung; Doh, Nakju
• Is Part Of: International journal of advanced robotic systems. Volume 16:Number 4(2019:Jul./Aug.)
• Publication Date: 2019-07-04
• Subjects: Normal estimation -- 3-D deep learning -- 3-D sensor system -- 3-D indoor LiDAR data set -- robustness; Robotics -- Periodicals; Robotics; Periodicals; 629.892
• DOI: 10.1177/1729881419857532
• Languages: English
• ISSNs: 1729-8806

Abstract

In this article, a point-wise normal estimation network for three-dimensional point cloud data called NormNet is proposed. We propose the multiscale K-nearest neighbor convolution module for strengthened local feature extraction. With the multiscale K-nearest neighbor convolution module and PointNet-like architecture, we achieved a hybrid of three features: a global feature, a semantic feature from the segmentation network, and a local feature from the multiscale K-nearest neighbor convolution module. Those features, by mutually supporting each other, not only increase the normal estimation performance but also enable the estimation to be robust under severe noise perturbations or point deficiencies. The performance was validated in three different data sets: Synthetic CAD data (ModelNet), RGB-D sensor-based real 3D PCD (S3DIS), and LiDAR sensor-based real 3D PCD that we built and shared.