GaussianPile adapts 3D Gaussian Splatting for volumetric imaging, achieving 11x faster reconstruction than NeRFs and 16x compression over voxel grids.
March 24, 2026
Original Paper
GaussianPile: A Unified Sparse Gaussian Splatting Framework for Slice-based Volumetric Reconstruction
arXiv · 2603.20611
The Takeaway
This brings the real-time rendering and storage benefits of Gaussian primitives to the medical and scientific imaging domain (ultrasound/microscopy), enabling fast 3D visualization on resource-constrained hardware.
From the abstract
Slice-based volumetric imaging is widely applied and it demands representations that compress aggressively while preserving internal structure for analysis. We introduce GaussianPile, unifying 3D Gaussian splatting with an imaging system-aware focus model to address this challenge. Our proposed method introduces three key innovations: (i) a slice-aware piling strategy that positions anisotropic 3D Gaussians to model through-slice contributions, (ii) a differentiable projection operator that enco