PERF: Performant, Explicit Radiance Fields

TL;DR

PERF introduces an explicit, non-neural radiance field approach for 3D reconstruction, enabling faster convergence and high-quality 360-degree scene reconstructions with reduced computation time.

Contribution

It presents a neural-network-free, explicit formulation of radiance fields using voxel grids and environment maps, improving convergence speed and reconstruction efficiency.

Findings

Achieves high-quality 3D reconstructions comparable to state-of-the-art methods.

Reduces reconstruction times significantly compared to neural network-based approaches.

Successfully reconstructs synthetic and real scenes from benchmark datasets.

Abstract

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{\deg} 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.