Reproduce “Robust 3DGS-based SLAM via Adaptive Kernel Smoothing” 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
Shouhe Zhang, Dayong Ren, Wen Jie Li, Piaopiao Yu, Sensen Song, Kaikai Shao, Yurong Qian
CB-KNN improves 3DGS-based SLAM tracking by temporarily correcting nearby Gaussian positions and colors during rendering, without changing map parameters. It adapts the correction to local density and motion. On TUM-RGBD, average ATE RMSE falls from 5.63 cm for SplaTAM to 3.67 cm. Averages also fall on Replica and ScanNet; these are benchmark results, not deployment evidence.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.