Data-driven Upsampling of Point Clouds

TL;DR

This paper introduces a deep learning-based method for upsampling sparse 3D point clouds, improving uniformity and accuracy without hard-coded rules, applicable across categories and outperforming traditional optimization methods.

Contribution

A novel data-driven deep network approach for 3D point cloud upsampling that learns from data and generalizes across object categories.

Findings

Outperforms baseline optimization-based methods in uniformity and accuracy.

Effective in both single-category and multi-category training scenarios.

Capable of handling various amplification factors.

Abstract

High quality upsampling of sparse 3D point clouds is critically useful for a wide range of geometric operations such as reconstruction, rendering, meshing, and analysis. In this paper, we propose a data-driven algorithm that enables an upsampling of 3D point clouds without the need for hard-coded rules. Our approach uses a deep network with Chamfer distance as the loss function, capable of learning the latent features in point clouds belonging to different object categories. We evaluate our algorithm across different amplification factors, with upsampling learned and performed on objects belonging to the same category as well as different categories. We also explore the desirable characteristics of input point clouds as a function of the distribution of the point samples. Finally, we demonstrate the performance of our algorithm in single-category training versus multi-category training scenarios. The final proposed model is compared against a baseline, optimization-based upsampling method. Results indicate that our algorithm is capable of generating more uniform and accurate upsamplings.