GURecon: Learning Detailed 3D Geometric Uncertainties for Neural Surface Reconstruction
This addresses a significant problem for researchers and practitioners in 3D reconstruction and neural rendering by providing a method to evaluate geometric quality without ground truth, though it is incremental as it builds on existing neural surface representations.
The paper tackles the challenge of assessing geometric quality in neural surface reconstructions without ground truth by introducing GURecon, a framework that learns a continuous 3D geometric uncertainty field based on geometric consistency, and experiments show its superiority in modeling uncertainty and improving downstream tasks like incremental reconstruction.
Neural surface representation has demonstrated remarkable success in the areas of novel view synthesis and 3D reconstruction. However, assessing the geometric quality of 3D reconstructions in the absence of ground truth mesh remains a significant challenge, due to its rendering-based optimization process and entangled learning of appearance and geometry with photometric losses. In this paper, we present a novel framework, i.e, GURecon, which establishes a geometric uncertainty field for the neural surface based on geometric consistency. Different from existing methods that rely on rendering-based measurement, GURecon models a continuous 3D uncertainty field for the reconstructed surface, and is learned by an online distillation approach without introducing real geometric information for supervision. Moreover, in order to mitigate the interference of illumination on geometric consistency, a decoupled field is learned and exploited to finetune the uncertainty field. Experiments on various datasets demonstrate the superiority of GURecon in modeling 3D geometric uncertainty, as well as its plug-and-play extension to various neural surface representations and improvement on downstream tasks such as incremental reconstruction. The code and supplementary material are available on the project website: https://zju3dv.github.io/GURecon/.