PyNeRF: Pyramidal Neural Radiance Fields
This addresses the problem of scale-aware rendering for accelerated NeRF methods, offering a simple, efficient solution for high-quality scene reconstruction, though it is incremental as it builds on existing grid-based approaches.
The paper tackles aliasing artifacts in accelerated Neural Radiance Fields (NeRFs) by proposing a pyramidal grid-based method that trains model heads at different resolutions, improving rendering quality with error reductions of 20-90% across scenes while maintaining minimal performance overhead.
Neural Radiance Fields (NeRFs) can be dramatically accelerated by spatial grid representations. However, they do not explicitly reason about scale and so introduce aliasing artifacts when reconstructing scenes captured at different camera distances. Mip-NeRF and its extensions propose scale-aware renderers that project volumetric frustums rather than point samples but such approaches rely on positional encodings that are not readily compatible with grid methods. We propose a simple modification to grid-based models by training model heads at different spatial grid resolutions. At render time, we simply use coarser grids to render samples that cover larger volumes. Our method can be easily applied to existing accelerated NeRF methods and significantly improves rendering quality (reducing error rates by 20-90% across synthetic and unbounded real-world scenes) while incurring minimal performance overhead (as each model head is quick to evaluate). Compared to Mip-NeRF, we reduce error rates by 20% while training over 60x faster.