NeRF-XL is a principled algorithm that enables the training and rendering of Neural Radiance Fields (NeRFs) with arbitrarily large capacity by efficiently distributing the NeRF parameters across multiple GPUs.
