ProbeSDF: Light Field Probes for Neural Surface Reconstruction
This work provides an incremental improvement for researchers and practitioners in 3D reconstruction by enhancing speed and accuracy in SDF-based methods.
The paper tackles the problem of slow and inefficient neural surface reconstruction by proposing a minimal radiance parametrization that decouples angular and spatial contributions, achieving faster training and improved performance in surface and image metrics on real data across diverse datasets.
SDF-based differential rendering frameworks have achieved state-of-the-art multiview 3D shape reconstruction. In this work, we re-examine this family of approaches by minimally reformulating its core appearance model in a way that simultaneously yields faster computation and increased performance. To this goal, we exhibit a physically-inspired minimal radiance parametrization decoupling angular and spatial contributions, by encoding them with a small number of features stored in two respective volumetric grids of different resolutions. Requiring as little as four parameters per voxel, and a tiny MLP call inside a single fully fused kernel, our approach allows to enhance performance with both surface and image (PSNR) metrics, while providing a significant training speedup and real-time rendering. We show this performance to be consistently achieved on real data over two widely different and popular application fields, generic object and human subject shape reconstruction, using four representative and challenging datasets.