Reproduce “Conditional-Flow NeRF: Accurate 3D Modelling with Reliable Uncertainty Quantification” 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

Jianxiong Shen, Antonio Agudo, Francesc Moreno-Noguer, Adria Ruiz

CF-NeRF models uncertainty in neural radiance fields by sampling a shared scene-level latent variable and using separate conditional normalizing flows for color and density. It produces scene samples and uncertainty estimates for rendered views and depth. On synthetic sparse-view benchmarks, it reports stronger image, depth, and uncertainty metrics across four complex Light Field scenes, but not across every dataset and metric.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.