RevolRecon: Neural Representation for Reconstructing Surface of Revolution

期刊论文
发表日期：

2025年5月21日

查看全文

Runqiao Li,
Qiujie Dong,
Shuangmin Chen

摘要

Neural signed distance functions (SDFs) have proved to be highly effective in reconstructing organic models. These neural SDFs are known for their expressiveness and ability to predict reasonable shapes even in the presence of incomplete data. However, when the target object is a surface of revolution, incorporating this prior into surface reconstruction remains a significant challenge. In this paper, we observe that for any point $\varvec{p}$ $p$ on a surface of revolution, the normal vector must align with the sectional plane passing through $\varvec{p}$ $p$ . Our approach, named RevolRecon, leverages this key observation in a self-supervised manner to handle missing data. Additionally, we propose a dynamic sampling strategy to extract points from the underlying surface, ensuring the loss function is estimated across the entire surface. Our approach also applies to the scenario of a curved rotation axis. A comprehensive comparison with state-of-the-art methods demonstrates the significant advantages of our approach.