Reproduce “Neural Stochastic Screened Poisson Reconstruction” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Silvia Sellán, Alec Jacobson
The authors present Neural Stochastic Screened Poisson Reconstruction, using neural networks to approximate a Gaussian-process prior over surfaces and uncertainty. That uncertainty guides scan selection, and new points update the reconstruction. On a mechanical object, it reduces Chamfer distance versus random views and matches or improves furthest-point sampling when views are sampled on a sphere and aimed at the same point.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.