HybridNeRF: Efficient Neural Rendering via Adaptive Volumetric Surfaces

arXiv:2312.03160v2 Announce Type: replace-cross
Abstract: Neural radiance fields provide state-of-the-art view synthesis quality but tend to be slow to render. One reason is that they make use of volume rendering, thus requiring many samples (and model queries) per ray at render time. Although this representation is flexible and easy to optimize, most real-world objects can be modeled more efficiently with surfaces instead of volumes, requiring far fewer samples per ray. This observation has spurred considerable progress in surface representations such …

abstract art arxiv cs.cv cs.gr cs.lg easy fields neural radiance fields neural rendering per quality queries ray reason rendering representation samples state synthesis type via view