VolRecon: Volume Rendering of Signed Ray Distance Functions for Generalizable Multi-View Reconstruction

The success of the Neural Radiance Fields (NeRF) in novel view synthesis has inspired researchers to propose neural implicit scene reconstruction. However, most existing neural implicit reconstruction methods optimize perscene parameters and therefore lack generalizability to new scenes. We introduce VolRecon, a novel generalizable implicit reconstruction method with Signed Ray Distance Function (SRDF). To reconstruct the scene with fine details and little noise, VolRecon combines projection features aggregated from multi-view features, and volume features interpolated from a coarse global feature volume. Using a ray transformer, we compute SRDF values of sampled points on a ray and then render color and depth. On DTU dataset, VolRecon outperforms SparseNeuS by about 30% in sparse view reconstruction and achieves comparable accuracy as MVSNet in full view reconstruction. Furthermore, our approach exhibits good generalization performance on the large-scale ETH3D benchmark. Code is available at https://github.com/IVRL/VolRecon/.

1. Introduction

The ability to reconstruct 3D geometries from images or videos is crucial in various applications in robotics [16], [43], [52] and augmented/virtual reality [29], [35]. Multi-view stereo (MVS) [13], [15], [39], [47], [54], [55] is a commonly used technique for this task. A typical MVS pipeline involves multiple steps, i.e., multi-view depth estimation, filtering, and fusion [5], [13].