CAP-UDF: Learning Unsigned Distance Functions Progressively from Raw Point Clouds with Consistency-Aware Field Optimization

TL;DR

CAP-UDF introduces a progressive, consistency-aware approach to learning unsigned distance functions from raw point clouds, enabling accurate surface reconstruction of both open and closed surfaces with improved smoothness and detail.

Contribution

The paper presents a novel method for learning UDFs that moves queries onto surfaces with a consistency constraint, improving surface reconstruction from raw point clouds.

Findings

Outperforms state-of-the-art methods in surface reconstruction accuracy.

Effectively reconstructs open and closed surfaces from raw point clouds.

Enhances unsupervised point normal estimation with the learned UDFs.

Abstract

Surface reconstruction for point clouds is an important task in 3D computer vision. Most of the latest methods resolve this problem by learning signed distance functions from point clouds, which are limited to reconstructing closed surfaces. Some other methods tried to represent open surfaces using unsigned distance functions (UDF) which are learned from ground truth distances. However, the learned UDF is hard to provide smooth distance fields due to the discontinuous character of point clouds. In this paper, we propose CAP-UDF, a novel method to learn consistency-aware UDF from raw point clouds. We achieve this by learning to move queries onto the surface with a field consistency constraint, where we also enable to progressively estimate a more accurate surface. Specifically, we train a neural network to gradually infer the relationship between queries and the approximated surface by searching for the moving target of queries in a dynamic way. Meanwhile, we introduce a polygonization algorithm to extract surfaces using the gradients of the learned UDF. We conduct comprehensive experiments in surface reconstruction for point clouds, real scans or depth maps, and further explore our performance in unsupervised point normal estimation, which demonstrate non-trivial improvements of CAP-UDF over the state-of-the-art methods.