HIVE: HIerarchical Volume Encoding for Neural Implicit Surface Reconstruction

TL;DR

This paper introduces a hierarchical volume encoding method to explicitly represent 3D shapes in neural implicit surface reconstruction, improving detail and smoothness over previous MLP-based approaches.

Contribution

The authors propose a novel hierarchical volume encoding that captures multi-scale spatial information, enhancing detail and smoothness in 3D reconstructions.

Findings

Significant improvement on DTU, EPFL, and BlendedMVS datasets.

Hierarchical volumes effectively encode multi-scale geometric details.

The method is compatible as a plug-and-play module with existing reconstruction techniques.

Abstract

Neural implicit surface reconstruction has become a new trend in reconstructing a detailed 3D shape from images. In previous methods, however, the 3D scene is only encoded by the MLPs which do not have an explicit 3D structure. To better represent 3D shapes, we introduce a volume encoding to explicitly encode the spatial information. We further design hierarchical volumes to encode the scene structures in multiple scales. The high-resolution volumes capture the high-frequency geometry details since spatially varying features could be learned from different 3D points, while the low-resolution volumes enforce the spatial consistency to keep the shape smooth since adjacent locations possess the same low-resolution feature. In addition, we adopt a sparse structure to reduce the memory consumption at high-resolution volumes, and two regularization terms to enhance results smoothness. This hierarchical volume encoding could be appended to any implicit surface reconstruction method as a plug-and-play module, and can generate a smooth and clean reconstruction with more details. Superior performance is demonstrated in DTU, EPFL, and BlendedMVS datasets with significant improvement on the standard metrics.