Point-NeRF: Point-based Neural Radiance Fields. (arXiv:2201.08845v1 [cs.CV])

Web: http://arxiv.org/abs/2201.08845

Volumetric neural rendering methods like NeRF generate high-quality view
synthesis results but are optimized per-scene leading to prohibitive
reconstruction time. On the other hand, deep multi-view stereo methods can
quickly reconstruct scene geometry via direct network inference. Point-NeRF
combines the advantages of these two approaches by using neural 3D point
clouds, with associated neural features, to model a radiance field. Point-NeRF
can be rendered efficiently by aggregating neural point features near scene
surfaces, in a ray marching-based rendering pipeline. Moreover, …

arxiv cv neural