$\texttt{GEM}$: 3D Gaussian Splatting for Efficient and Accurate Cryo-EM Reconstruction

Huaizhi Qu, Xiao Wang, Gengwei Zhang, Jie Peng, Tianlong Chen

Keywords: 3D Reconstruction, Cryo-EM, 3D Gaussian Splatting

TL;DR: We develop a 3DGS-based cryo-EM reconstruction method that achieves high accuracy and efficiency

Abstract: Cryo-electron microscopy (cryo-EM) has become a central tool for high-resolution structural biology, yet the massive scale of datasets (often exceeding 100k particle images) renders 3D reconstruction both computationally expensive and memory intensive. Traditional Fourier-space methods are efficient but lose fidelity due to repeated transforms, while recent real-space approaches based on neural radiance fields (NeRFs) improve accuracy but incur cubic memory and computation overhead. We introduce $\texttt{GEM}$, a novel cryo-EM reconstruction framework built on 3D Gaussian Splatting (3DGS) that operates directly in real-space while maintaining high efficiency. Instead of modeling the entire density volume, $\texttt{GEM}$ represents proteins with hundreds of thousands of compact 3D Gaussians, each parameterized by only 11 values. An efficient implementation further restricts gradient computation to 3D Gaussians that contribute to each pixel, substantially reducing both memory footprint and training cost. On standard cryo-EM benchmarks, $\texttt{GEM}$ achieves up to $48\times$ faster training and $12\times$ lower memory usage compared to state-of-the-art methods, while improving local resolution by as much as $38.8\\%$. These results establish $\texttt{GEM}$ as a practical and scalable paradigm for cryo-EM reconstruction, unifying speed, efficiency, and high-resolution accuracy.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Submission Number: 6298