Sparse convolutional networks for surface reconstruction from noisy point clouds

Wang, Tao, Wu, Jing ORCID: https://orcid.org/0000-0001-5123-9861, Ji, Ze ORCID: https://orcid.org/0000-0002-8968-9902 and Lai, Yukun ORCID: https://orcid.org/0000-0002-2094-5680 2024. Sparse convolutional networks for surface reconstruction from noisy point clouds. Presented at: IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Hawaii, USA, 4-8 January 2024. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. IEEE, pp. 3200-3209. 10.1109/WACV57701.2024.00318

Preview

PDF - Accepted Post-Print Version Download (2MB) | Preview

Abstract

Reconstructing accurate 3D surfaces from noisy point clouds is a fundamental problem in computer vision. Among different approaches, neural implicit methods that map 3D coordinates to occupancy values benefit from the learning capabilities of deep neural networks and the flexible topology of implicit representations, achieving promising reconstruction results. However, existing methods utilize standard (dense) 3D convolutional neural networks for feature extraction and occupancy prediction, which significantly restricts their capability to reconstruct details. In this paper, we propose a neural implicit method based on sparse convolutions, where features and network calculations only focus on grid points close to the surface to be reconstructed. This allows us to build significantly higher resolution 3D grids and reconstruct high-fidelity details. We further build a 3D residual UNet to extract features which are robust to noise, while ensuring details are retained. A 3D position along with features extracted at the position are fed into the occupancy probability predictor network to obtain occupancy. As features at nearby grid points to the query position may not exist due to the sparse nature, we propose a normalized weight interpolation approach to obtain smooth interpolation with sparse data. Experimental results demonstrate that our method achieves promising results, both qualitatively and quantitatively, outperforming existing methods.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Professional Services > Advanced Research Computing @ Cardiff (ARCCA) Schools > Computer Science & Informatics Schools > Engineering
Publisher:	IEEE
ISBN:	979-8-3503-1893-7
ISSN:	2472-6737
Date of First Compliant Deposit:	9 November 2023
Date of Acceptance:	24 October 2023
Last Modified:	12 Mar 2025 14:30
URI:	https://orca.cardiff.ac.uk/id/eprint/163775

Actions (repository staff only)

Edit Item