Publications
Detailed Information
Deep Point Cloud Reconstruction
Cited 0 time in
Web of Science
Cited 4 time in Scopus
- Authors
- Issue Date
- 2022
- Citation
- ICLR 2022 - 10th International Conference on Learning Representations
- Abstract
- Point cloud obtained from 3D scanning is often sparse, noisy, and irregular. To cope with these issues, recent studies have been separately conducted to densify, denoise, and complete inaccurate point cloud. In this paper, we advocate that jointly solving these tasks leads to significant improvement for point cloud reconstruction. To this end, we propose a deep point cloud reconstruction network consisting of two stages: 1) a 3D sparse stacked-hourglass network as for the initial densification and denoising, 2) a refinement via transformers converting the discrete voxels into 3D points. In particular, we further improve the performance of transformer by a newly proposed module called amplified positional encoding. This module has been designed to differently amplify the magnitude of the positional encoding vectors based on the points' distances. Extensive experiments demonstrate that our network achieves state-of-the-art performance among the recent studies in the ScanNet, ICL-NUIM, and ShapeNetPart datasets. Moreover, we underline the ability of our network to generalize toward real-world and unmet scenes.
- Files in This Item:
- There are no files associated with this item.
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.