Reproduce “Spiking GS: Towards High-Accuracy and Low-Cost Surface Reconstruction via Spiking Neuron-based Gaussian Splatting” 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

Weixing Zhang, Zongrui Li, De Ma, Huajin Tang, Xudong Jiang, Qian Zheng, Gang Pan

Spiking GS improves 3D surface reconstruction from multiview photos by suppressing low-opacity Gaussians and faint tails within Gaussians before extracting a mesh from depth estimates. Across eight NeRF-Synthetic scenes, it reports average Chamfer distance 0.87 × 10⁻², versus 1.04 × 10⁻² for 2DGS, using 69k rather than 133k Gaussians. Its advantage diminishes on DTU, and thin structures remain challenging.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.