# Real-time coherent diffraction inversion using deep generative networks

arXiv: https://arxiv.org/abs/1806.03992 · HTML: https://mlsanity.com/paper/1806.03992/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Jun 2018.

## Built on by (most cited first, 17 of 17)

- [On the use of deep learning for phase recovery](https://mlsanity.com/paper/2308.00942/) (2023)
- [AutoPhaseNN: Unsupervised Physics-aware Deep Learning of 3D Nanoscale Bragg Coherent Diffraction Imaging](https://mlsanity.com/paper/2109.14053/) (2021)
- [Practical Phase Retrieval Using Double Deep Image Priors](https://mlsanity.com/paper/2211.00799/) (2022)
- [Compressive Ptychography using Deep Image and Generative Priors](https://mlsanity.com/paper/2205.02397/) (2022)
- [Deep-learning real-time phase retrieval of imperfect diffraction patterns from X-ray free-electron lasers](https://mlsanity.com/paper/2409.15784/) (2024)
- [Neural Network Methods for Radiation Detectors and Imaging](https://mlsanity.com/paper/2311.05726/) (2023)
- [What is Wrong with End-to-End Learning for Phase Retrieval?](https://mlsanity.com/paper/2403.15448/) (2024)
- [Amortized Bayesian Inference of GISAXS Data with Normalizing Flows](https://mlsanity.com/paper/2210.01543/) (2022)
- [Investigating the robustness of a learning-based method for quantitative phase retrieval from propagation-based x-ray phase contrast measurements under laboratory conditions](https://mlsanity.com/paper/2211.01372/) (2022)
- [AI-assisted Automated Workflow for Real-time X-ray Ptychography Data Analysis via Federated Resources](https://mlsanity.com/paper/2304.04297/) (2023)
- [Deep learning phase recovery: data-driven, physics-driven, or combining both?](https://mlsanity.com/paper/2404.01360/) (2024)
- [DONUT: Physics-aware Machine Learning for Real-time X-ray Nanodiffraction Analysis](https://mlsanity.com/paper/2507.14038/) (2025)
- [Position-Blind Ptychography: Viability of image reconstruction via data-driven variational inference](https://mlsanity.com/paper/2509.25269/) (2025)
- [Circular Phase Representation and Geometry-Aware Optimization for Ptychographic Image Reconstruction](https://mlsanity.com/paper/2604.26664/) (2026)
- [A Deep Generative Approach to Oversampling in Ptychography](https://mlsanity.com/paper/2207.14392/) (2022)
- [Coordinate-based Neural Network for Fourier Phase Retrieval](https://mlsanity.com/paper/2311.14925/) (2023)
- [A unified self-supervised framework for single-frame Fresnel CDI and overlapped ptychography](https://mlsanity.com/paper/2602.21361/) (2026)

## Abstract (arXiv)

Phase retrieval, or the process of recovering phase information in reciprocal space to reconstruct images from measured intensity alone, is the underlying basis to a variety of imaging applications including coherent diffraction imaging (CDI). Typical phase retrieval algorithms are iterative in nature, and hence, are time-consuming and computationally expensive, precluding real-time imaging. Furthermore, iterative phase retrieval algorithms struggle to converge to the correct solution especially in the presence of strong phase structures. In this work, we demonstrate the training and testing of CDI NN, a pair of deep deconvolutional networks trained to predict structure and phase in real space of a 2D object from its corresponding far-field diffraction intensities alone. Once trained, CDI NN can invert a diffraction pattern to an image within a few milliseconds of compute time on a standard desktop machine, opening the door to real-time imaging.