PERF: Performant, Explicit Radiance Fields
This addresses the problem of slow reconstruction times in radiance field methods for 3D reconstruction, offering a performant alternative for applications requiring efficient scene modeling.
The paper tackles 3D reconstruction from images by formulating volumetric reconstruction as a non-linear least-squares problem solved explicitly without neural networks, achieving quality on par with state-of-the-art methods with significantly reduced reconstruction times.
We present a novel way of approaching image-based 3D reconstruction based on radiance fields. The problem of volumetric reconstruction is formulated as a non-linear least-squares problem and solved explicitly without the use of neural networks. This enables the use of solvers with a higher rate of convergence than what is typically used for neural networks, and fewer iterations are required until convergence. The volume is represented using a grid of voxels, with the scene surrounded by a hierarchy of environment maps. This makes it possible to get clean reconstructions of 360° scenes where the foreground and background is separated. A number of synthetic and real scenes from well known benchmark-suites are successfully reconstructed with quality on par with state-of-the-art methods, but at significantly reduced reconstruction times.