✳flâneur — a map of the web's best reading

Zip-NeRF

jonbarron.info · 498 words · saved by 1 readers

Neural Radiance Field training can be accelerated through the use of grid-based representations in NeRF's learned mapping from spatial coordinates to colors and volumetric density. However, these grid-based approaches lack an explicit understanding of scale and therefore often introduce aliasing, usually in the form of jaggies or missing scene content. Anti-aliasing has previously been addressed by mip-NeRF 360, which reasons about sub-volumes along a cone rather than points along a ray, but this approach is not natively compatible with current grid-based techniques. We show how ideas from rendering and signal processing can be used to construct a technique that combines mip-NeRF 360 and grid-based models such as Instant NGP to yield error rates that are 8%-77% lower than either prior technique, and that trains 24x faster than mip-NeRF 360. We use multisampling to approximate the average NGP feature over a conical frustum, by constructing a 6-sample pattern that exactly matches the fru

Zip-NeRF Zip-NeRF : Anti-Aliased Grid-Based Neural Radiance Fields ICCV 2023 (Oral Presentation, Best Paper Finalist) Jonathan T. Barron Ben Mildenhall Dor Verbin Pratul P. Srinivasan Peter Hedman Google Paper Video Code Data Google Vertex AI Notebook Abstract Neural Radiance Field training can be accelerated through the use of grid-based representations in NeRF's learned mapping from spatial coordinates to colors and volumetric density. However, these grid-based approaches lack an explicit understanding of scale and therefore often introduce aliasing, usually in the form of jaggies or missing

Explore this link on the map →

related reading