Sphere-Guided Training of Neural Implicit Surfaces
This work addresses a bottleneck in multi-view 3D reconstruction for computer vision applications, offering an incremental improvement over existing methods.
The paper tackles the inefficiency of volumetric ray marching in neural implicit surfaces for 3D reconstruction by introducing a coarse sphere-based representation to exclude empty volumes, leading to improved reconstruction fidelity across synthetic and real-world datasets.
In recent years, neural distance functions trained via volumetric ray marching have been widely adopted for multi-view 3D reconstruction. These methods, however, apply the ray marching procedure for the entire scene volume, leading to reduced sampling efficiency and, as a result, lower reconstruction quality in the areas of high-frequency details. In this work, we address this problem via joint training of the implicit function and our new coarse sphere-based surface reconstruction. We use the coarse representation to efficiently exclude the empty volume of the scene from the volumetric ray marching procedure without additional forward passes of the neural surface network, which leads to an increased fidelity of the reconstructions compared to the base systems. We evaluate our approach by incorporating it into the training procedures of several implicit surface modeling methods and observe uniform improvements across both synthetic and real-world datasets. Our codebase can be accessed via the project page: https://andreeadogaru.github.io/SphereGuided