Improving Multislice Electron Ptychography with a Generative Prior
This work addresses reconstruction quality for researchers in materials science and electron microscopy, but it is incremental as it augments existing methods rather than introducing a new paradigm.
The authors tackled the problem of suboptimal and time-consuming reconstructions in multislice electron ptychography by integrating a diffusion model as a generative prior into existing iterative solvers, achieving a 90.50% improvement in SSIM.
Multislice electron ptychography (MEP) is an inverse imaging technique that computationally reconstructs the highest-resolution images of atomic crystal structures from diffraction patterns. Available algorithms often solve this inverse problem iteratively but are both time consuming and produce suboptimal solutions due to their ill-posed nature. We develop MEP-Diffusion, a diffusion model trained on a large database of crystal structures specifically for MEP to augment existing iterative solvers. MEP-Diffusion is easily integrated as a generative prior into existing reconstruction methods via Diffusion Posterior Sampling (DPS). We find that this hybrid approach greatly enhances the quality of the reconstructed 3D volumes, achieving a 90.50% improvement in SSIM over existing methods.