ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume IV-2/W5
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-2/W5, 307–312, 2019
https://doi.org/10.5194/isprs-annals-IV-2-W5-307-2019
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-2/W5, 307–312, 2019
https://doi.org/10.5194/isprs-annals-IV-2-W5-307-2019

  29 May 2019

29 May 2019

INDOOR SCENE REGISTRATION BASED ON SIAMESE NETWORK AND POINTNET

Z. Zhang1, C. Wen1, Y. Chen1, W. Li1, C. You1, C. Wang1, and J. Li1,2 Z. Zhang et al.
  • 1Fujian Key Laboratory of Sensing and Computing for Smart Cities, Xiamen University, Xiamen, China
  • 2Department of Geography and Environmental Management, University of Waterloo, Waterloo, Canada

Keywords: Mobile LiDAR Data, Registration, Descriptor, Siamese Network, PointNet, Indoor Scene

Abstract. This paper presents a deep learning feature-based method for registration of indoor mobile LiDAR data. Our method is to input point cloud directly, which is more robust to noise than traditional algorithms. The proposed method involves three steps. We first extract the key points by Harris3D algorithm and get their local patches by our sampling method. Second, a Siamese network is trained to describe the patches as local descriptors. Finally, we obtain the final matching pairs depends on the distance which is between two descriptors, and then solve the transformation matrix. The accuracy of registration is within 6 cm when the overlap is greater than 35%. In order to improve the registration accuracy, the ICP algorithm is used to fine-tuning the registration results. And the final registration accuracy is within 3.5 cm. The experiments show that our method applied to the registration of indoor mobile LiDAR data robustly and accurately.