Reproduce “RPC-GS: Gaussian Splatting with native RPC Rendering for Satellite Imagery” 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
Valentin Wagner, Sebastian Bullinger, Christoph Bodensteiner, Michael Arens
RPC-GS uses satellite Rational Polynomial Camera (RPC) geometry directly in a Gaussian-splatting renderer to reconstruct 3D terrain from overlapping images. It transforms scene coordinates and projects Gaussian covariances through the RPC model, while ray-based metric depth orders splats. Across four DFC2019 and three IARPA2016 scenes, it has the lowest mean altitude error but lower mean PSNR; results compare camera-and-renderer configurations.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.