PERF: Performant, Explicit Radiance Fields

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° 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.

CP‐NeRF: Conditionally Parameterized Neural Radiance Fields for Cross‐scene Novel View Synthesis

“…InstanceNGP [MESK22] deploys compact multilayer perceptrons (MLPs) in tandem with hash‐encoding to achieve more efficient training. Similarly, PERF [RSA22] employs Gauss‐Newton approximations as second‐order derivatives, providing an alternative to traditional optimizers like Adam. However, it should be emphasized that the scope of these methods is generally limited to optimization on single scenes.…”