HelixSurf: A Robust and Efficient Neural Implicit Surface Learning of Indoor Scenes with Iterative Intertwined Regularization

TL;DR

HelixSurf introduces an iterative intertwined regularization approach combining neural implicit surface learning and multi-view stereo to improve indoor scene reconstruction, achieving faster results with better accuracy.

Contribution

The paper proposes HelixSurf, a novel method that integrates neural implicit surface learning with multi-view stereo through iterative regularization for enhanced indoor scene reconstruction.

Findings

Outperforms existing methods in indoor scene surface reconstruction.

Significantly faster than comparable approaches, even with auxiliary data.

Effective regularization improves surface detail and accuracy.

Abstract

Recovery of an underlying scene geometry from multiview images stands as a long-time challenge in computer vision research. The recent promise leverages neural implicit surface learning and differentiable volume rendering, and achieves both the recovery of scene geometry and synthesis of novel views, where deep priors of neural models are used as an inductive smoothness bias. While promising for object-level surfaces, these methods suffer when coping with complex scene surfaces. In the meanwhile, traditional multi-view stereo can recover the geometry of scenes with rich textures, by globally optimizing the local, pixel-wise correspondences across multiple views. We are thus motivated to make use of the complementary benefits from the two strategies, and propose a method termed Helix-shaped neural implicit Surface learning or HelixSurf; HelixSurf uses the intermediate prediction from one strategy as the guidance to regularize the learning of the other one, and conducts such intertwined regularization iteratively during the learning process. We also propose an efficient scheme for differentiable volume rendering in HelixSurf. Experiments on surface reconstruction of indoor scenes show that our method compares favorably with existing methods and is orders of magnitude faster, even when some of existing methods are assisted with auxiliary training data. The source code is available at https://github.com/Gorilla-Lab-SCUT/HelixSurf.