PU-Net: Point Cloud Upsampling Network

• Published in: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 2790 - 2799
• Main Authors: Yu, Lequan, Li, Xianzhi, Fu, Chi-Wing, Cohen-Or, Daniel, Heng, Pheng-Ann
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2018
• Subjects: Convolution; Feature extraction; Geometry; Image reconstruction; Surface reconstruction; Three-dimensional displays; Training
• ISSN: 1063-6919
• DOI: 10.1109/CVPR.2018.00295

Learning and analyzing 3D point clouds with deep networks is challenging due to the sparseness and irregularity of the data. In this paper, we present a data-driven point cloud upsampling technique. The key idea is to learn multi-level features per point and expand the point set via a multi-branch convolution unit implicitly in feature space. The expanded feature is then split to a multitude of features, which are then reconstructed to an upsampled point set. Our network is applied at a patch-level, with a joint loss function that encourages the upsampled points to remain on the underlying surface with a uniform distribution. We conduct various experiments using synthesis and scan data to evaluate our method and demonstrate its superiority over some baseline methods and an optimization-based method. Results show that our upsampled points have better uniformity and are located closer to the underlying surfaces.