NeAT: Neural Adaptive Tomography
This addresses the challenge of efficient and high-quality tomography reconstruction for semi-transparent volumetric scenes, representing an incremental improvement over existing neural and optimization-based methods.
The paper tackles the problem of multi-view inverse rendering for semi-transparent volumetric scenes by introducing Neural Adaptive Tomography (NeAT), achieving reconstruction times far superior to existing neural methods and outperforming optimization-based solvers in quality while being substantially faster.
In this paper, we present Neural Adaptive Tomography (NeAT), the first adaptive, hierarchical neural rendering pipeline for multi-view inverse rendering. Through a combination of neural features with an adaptive explicit representation, we achieve reconstruction times far superior to existing neural inverse rendering methods. The adaptive explicit representation improves efficiency by facilitating empty space culling and concentrating samples in complex regions, while the neural features act as a neural regularizer for the 3D reconstruction. The NeAT framework is designed specifically for the tomographic setting, which consists only of semi-transparent volumetric scenes instead of opaque objects. In this setting, NeAT outperforms the quality of existing optimization-based tomography solvers while being substantially faster.