Reproduce “OUGS: Active View Selection via Object-aware Uncertainty Estimation in 3DGS” 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
Haiyi Li, Qi Chen, Denis Kalkofen, Hsiang-Ting Chen
OUGS selects views for 3D reconstruction by estimating uncertainty in Gaussian parameters, projecting it into image space, and weighting view scores with an object mask. On Mip-NeRF360, at a matched budget of 20 views, it reports object-mask PSNR of 29.6099, versus 27.3012 for GauSS-MI. This improves target fidelity, not performance universally, and refines represented regions rather than discovering missing structure.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.