# Few-shot Transfer Learning for Holographic Image Reconstruction using a Recurrent Neural Network

arXiv: https://arxiv.org/abs/2201.11333 · HTML: https://mlsanity.com/paper/2201.11333/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 40 references that are in the map.
- First posted on arXiv 27 Jan 2022.

## Builds on (most cited first, 13 of 13)

- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Adversarial Discriminative Domain Adaptation](https://mlsanity.com/paper/1702.05464/) (2017)
- [Least Squares Generative Adversarial Networks](https://mlsanity.com/paper/1611.04076/) (2016)
- [Deep learning-based virtual histology staining using auto-fluorescence of label-free tissue](https://mlsanity.com/paper/1803.11293/) (2018)
- [Deep learning-based transformation of the H&E stain into special stains](https://mlsanity.com/paper/2008.08871/) (2020)
- [Phase recovery and holographic image reconstruction using deep learning in neural networks](https://mlsanity.com/paper/1705.04286/) (2017)
- [Low Photon Count Phase Retrieval Using Deep Learning](https://mlsanity.com/paper/1806.10029/) (2018)
- [Deep Phase Decoder: Self-calibrating phase microscopy with an untrained deep neural network](https://mlsanity.com/paper/2001.09803/) (2020)
- [Cross-modality deep learning brings bright-field microscopy contrast to holography](https://mlsanity.com/paper/1811.07103/) (2018)
- [Learning to Synthesize: Robust Phase Retrieval at Low Photon counts](https://mlsanity.com/paper/1907.11713/) (2019)
- [Holographic image reconstruction with phase recovery and autofocusing using recurrent neural networks](https://mlsanity.com/paper/2102.12281/) (2021)
- [Recurrent neural network-based volumetric fluorescence microscopy](https://mlsanity.com/paper/2010.10781/) (2020)
- [Deep learning-based virtual refocusing of images using an engineered point-spread function](https://mlsanity.com/paper/2012.11892/) (2020)

## Built on by (most cited first, 7 of 7)

- [Fourier Imager Network (FIN): A deep neural network for hologram reconstruction with superior external generalization](https://mlsanity.com/paper/2204.10533/) (2022)
- [Self-supervised learning of hologram reconstruction using physics consistency](https://mlsanity.com/paper/2209.08288/) (2022)
- [Virtual stain transfer in histology via cascaded deep neural networks](https://mlsanity.com/paper/2207.06578/) (2022)
- [Domain-Aware Few-Shot Learning for Optical Coherence Tomography Noise Reduction](https://mlsanity.com/paper/2306.08102/) (2023)
- [Neural Network-Based Processing and Reconstruction of Compromised Biophotonic Image Data](https://mlsanity.com/paper/2403.14324/) (2024)
- [One-Shot Image Restoration](https://mlsanity.com/paper/2404.17426/) (2024)
- [Saving Foundation Flow-Matching Priors for Inverse Problems](https://mlsanity.com/paper/2511.16520/) (2025)

## Abstract (arXiv)

Deep learning-based methods in computational microscopy have been shown to be powerful but in general face some challenges due to limited generalization to new types of samples and requirements for large and diverse training data. Here, we demonstrate a few-shot transfer learning method that helps a holographic image reconstruction deep neural network rapidly generalize to new types of samples using small datasets. We pre-trained a convolutional recurrent neural network on a large dataset with diverse types of samples, which serves as the backbone model. By fixing the recurrent blocks and transferring the rest of the convolutional blocks of the pre-trained model, we reduced the number of trainable parameters by ~90% compared with standard transfer learning, while achieving equivalent generalization. We validated the effectiveness of this approach by successfully generalizing to new types of samples using small holographic datasets for training, and achieved (i) ~2.5-fold convergence speed acceleration, (ii) ~20% computation time reduction per epoch, and (iii) improved reconstruction performance over baseline network models trained from scratch. This few-shot transfer learning approach can potentially be applied in other microscopic imaging methods, helping to generalize to new types of samples without the need for extensive training time and data.