MCNeRF: Monte Carlo Rendering and Denoising for Real-Time NeRFs

1University of California San Diego, 2Adobe Research

MCNeRF is a general Monte Carlo-based method to accelerate the rendering of any NeRF representation. It boosts the speed of TensoRF, a vanilla NeRF approach, by 7x with minimal quality loss. In contrast, MobileNeRF, a state-of-the-art method, sacrifices complex fuzzy appearance for real-time performance by baking the NeRF model onto a mesh.


The volume rendering step used in Neural Radiance Fields (NeRFs) produces highly photorealistic results, but is inherently slow because it evaluates an MLP at a large number of sample points per ray.

Previous work has addressed this by either proposing neural scene representations that are faster to evaluate or by pre-computing (and approximating) scene properties to reduce render times. In this work, we propose MCNeRF, a general Monte Carlo-based rendering algorithm that can speed up any NeRF representation.

We show that the NeRF volume rendering integral can be efficiently computed via Monte Carlo integration using an importance sampling scheme based on ray density distributions. This allows us to use a small number of MLP evaluations to estimate pixel radiance. These noisy Monte Carlo estimates can be further denoised using an inexpensive image-space denoiser trained per-scene. We demonstrate that MCNeRF can be used to speed up NeRF representations like TensoRF by 7× while closely matching their visual quality and without making the scene approximations that real-time NeRF rendering methods usually make.