SuperUDF: Self-supervised UDF Estimation for Surface Reconstruction

TL;DR

SuperUDF introduces a self-supervised learning method for unsigned distance functions that improves surface reconstruction quality and efficiency, especially with sparse data, by leveraging a geometry prior and novel regularization techniques.

Contribution

It presents a novel self-supervised UDF learning framework with a geometry prior and regularization, enabling robust and efficient surface reconstruction from sparse samples.

Findings

Outperforms state-of-the-art methods in quality and efficiency

Robust to sparse sampling due to new regularization

Provides a learning-based mesh extraction method

Abstract

Learning-based surface reconstruction based on unsigned distance functions (UDF) has many advantages such as handling open surfaces. We propose SuperUDF, a self-supervised UDF learning which exploits a learned geometry prior for efficient training and a novel regularization for robustness to sparse sampling. The core idea of SuperUDF draws inspiration from the classical surface approximation operator of locally optimal projection (LOP). The key insight is that if the UDF is estimated correctly, the 3D points should be locally projected onto the underlying surface following the gradient of the UDF. Based on that, a number of inductive biases on UDF geometry and a pre-learned geometry prior are devised to learn UDF estimation efficiently. A novel regularization loss is proposed to make SuperUDF robust to sparse sampling. Furthermore, we also contribute a learning-based mesh extraction from the estimated UDFs. Extensive evaluations demonstrate that SuperUDF outperforms the state of the arts on several public datasets in terms of both quality and efficiency. Code url is https://github.com/THHHomas/SuperUDF.