Reproduce “GSSA: Gaussian Surfels with Spatial Awareness for Surface Reconstruction” 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
Haojun Tang, Siyuan Zou, Hongbo Pan, Yixin Lu, Shun Zhou
GSSA extracts meshes from trained Gaussian Surfels by using their 3D geometry to build a signed distance field, rather than fusing depth images rendered from multiple views. It refines surfels and combines nearby distance estimates in a voxel field. On DTU, F1 is 84.5%, versus 82.3% for PGSR. Results vary across benchmarks, and incomplete surfel coverage can leave regions missing.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.