Reproduce “AffineGS: Unlocking Accurate Geometry in 3DGS via Spatially Varying Affine Rectification” 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
Tze Ho Elden Tse, Jizong Peng, Richard Chen, Angela Yao
AffineGS uses locally varying affine corrections to align monocular depth guidance with 3D Gaussian Splatting (3DGS), then couples corrected guidance to depth gradients and surface normals. On MuSHRoom, its mesh F-score is 0.7097 versus 0.2765 for a globally aligned monocular-depth baseline, within training-camera-visible regions; sensor depths initialize indoor scenes, and the method does not establish physical scale.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.