Convolutional Neural Network-based Efficient Dense Point Cloud Generation using Unsigned Distance Fields

TL;DR

This paper introduces a lightweight CNN that efficiently generates dense 3D point clouds from sparse data by predicting unsigned distance fields, outperforming existing methods in speed, size, and quality.

Contribution

The authors propose a novel CNN architecture that learns unsigned distance fields for arbitrary shapes, enabling efficient and high-quality dense point cloud generation.

Findings

7.8x fewer model parameters

2.4x faster inference time

up to 24.8% better generation quality

Abstract

Dense point cloud generation from a sparse or incomplete point cloud is a crucial and challenging problem in 3D computer vision and computer graphics. So far, the existing methods are either computationally too expensive, suffer from limited resolution, or both. In addition, some methods are strictly limited to watertight surfaces -- another major obstacle for a number of applications. To address these issues, we propose a lightweight Convolutional Neural Network that learns and predicts the unsigned distance field for arbitrary 3D shapes for dense point cloud generation using the recently emerged concept of implicit function learning. Experiments demonstrate that the proposed architecture outperforms the state of the art by 7.8x less model parameters, 2.4x faster inference time and up to 24.8% improved generation quality compared to the state-of-the-art.