# Enhancing Digital Hologram Reconstruction Using Reverse-Attention Loss for Untrained Physics-Driven Deep Learning Models with Uncertain Distance

arXiv: https://arxiv.org/abs/2403.12056 · HTML: https://mlsanity.com/paper/2403.12056/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 42 references that are in the map.
- First posted on arXiv 11 Jan 2024.

## Builds on (most cited first, 14 of 14)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Convex optimization](https://mlsanity.com/paper/2106.01946/) (2021)
- [Deep Image Prior](https://mlsanity.com/paper/1711.10925/) (2017)
- [Lookahead Optimizer: k steps forward, 1 step back](https://mlsanity.com/paper/1907.08610/) (2019)
- [Phase recovery and holographic image reconstruction using deep learning in neural networks](https://mlsanity.com/paper/1705.04286/) (2017)
- [Fourier Imager Network (FIN): A deep neural network for hologram reconstruction with superior external generalization](https://mlsanity.com/paper/2204.10533/) (2022)
- [Cross-modality deep learning brings bright-field microscopy contrast to holography](https://mlsanity.com/paper/1811.07103/) (2018)
- [Convolutional Neural Network (CNN) vs Vision Transformer (ViT) for Digital Holography](https://mlsanity.com/paper/2108.09147/) (2021)
- [Deep DIH : Statistically Inferred Reconstruction of Digital In-Line Holography by Deep Learning](https://mlsanity.com/paper/2004.12231/) (2020)
- [Fast Key Points Detection and Matching for Tree-Structured Images](https://mlsanity.com/paper/2211.03242/) (2022)
- [Convolutional neural network-based regression for depth prediction in digital holography](https://mlsanity.com/paper/1802.00664/) (2018)
- [Fast Autofocusing using Tiny Transformer Networks for Digital Holographic Microscopy](https://mlsanity.com/paper/2203.07772/) (2022)
- [Nano-Resolution Visual Identifiers Enable Secure Monitoring in Next-Generation Cyber-Physical Systems](https://mlsanity.com/paper/2211.08678/) (2022)
- [Randomness assisted in-line holography with deep learning](https://mlsanity.com/paper/2303.17660/) (2023)

## Built on by (most cited first, 1 of 1)

- [Diffusion Prism: Enhancing Diversity and Morphology Consistency in Mask-to-Image Diffusion](https://mlsanity.com/paper/2501.00944/) (2025)

## Abstract (arXiv)

Untrained Physics-based Deep Learning (DL) methods for digital holography have gained significant attention due to their benefits, such as not requiring an annotated training dataset, and providing interpretability since utilizing the governing laws of hologram formation. However, they are sensitive to the hard-to-obtain precise object distance from the imaging plane, posing the textitAutofocusing challenge. Conventional solutions involve reconstructing image stacks for different potential distances and applying focus metrics to select the best results, which apparently is computationally inefficient. In contrast, recently developed DL-based methods treat it as a supervised task, which again needs annotated data and lacks generalizability. To address this issue, we propose textitreverse-attention loss, a weighted sum of losses for all possible candidates with learnable weights. This is a pioneering approach to addressing the Autofocusing challenge in untrained deep-learning methods. Both theoretical analysis and experiments demonstrate its superiority in efficiency and accuracy. Interestingly, our method presents a significant reconstruction performance over rival methods (i.e. alternating descent-like optimization, non-weighted loss integration, and random distance assignment) and even is almost equal to that achieved with a precisely known object distance. For example, the difference is less than 1dB in PSNR and 0.002 in SSIM for the target sample in our experiment.