Abstract | ||
---|---|---|
Volumetric neural rendering methods like NeRF [34] 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, Point-NeRF can be initialized via direct inference of a pre-trained deep network to produce a neural point cloud; this point cloud can be finetuned to surpass the visual quality of NeRF with 30× faster training time. Point-NeRF can be combined with other 3D re-construction methods and handles the errors and outliers in such methods via a novel pruning and growing mechanism. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/CVPR52688.2022.00536 | IEEE Conference on Computer Vision and Pattern Recognition |
Keywords | DocType | Volume |
3D from multi-view and sensors, Computational photography, Image and video synthesis and generation | Conference | 2022 |
Issue | Citations | PageRank |
1 | 0 | 0.34 |
References | Authors | |
0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Qiangeng Xu | 1 | 20 | 2.61 |
Zexiang Xu | 2 | 101 | 10.17 |
Julien Philip | 3 | 0 | 0.34 |
Sai Bi | 4 | 63 | 5.28 |
Zhixin Shu | 5 | 13 | 5.26 |
Kalyan Sunkavalli | 6 | 500 | 31.75 |
Ulrich Neumann | 7 | 2218 | 191.28 |