# Unfolding Framework with Complex-Valued Deformable Attention for High-Quality Computer-Generated Hologram Generation

arXiv: https://arxiv.org/abs/2508.21657 · HTML: https://mlsanity.com/paper/2508.21657/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 35 references that are in the map.
- First posted on arXiv 29 Aug 2025.

## Builds on (most cited first, 9 of 9)

- [Enhanced Deep Residual Networks for Single Image Super-Resolution](https://mlsanity.com/paper/1707.02921/) (2017)
- [Vision Transformer with Deformable Attention](https://mlsanity.com/paper/2201.00520/) (2022)
- [Phase Retrieval via Wirtinger Flow: Theory and Algorithms](https://mlsanity.com/paper/1407.1065/) (2014)
- [Embedding Fourier for Ultra-High-Definition Low-Light Image Enhancement](https://mlsanity.com/paper/2302.11831/) (2023)
- [Gradient Descent with Random Initialization: Fast Global Convergence for Nonconvex Phase Retrieval](https://mlsanity.com/paper/1803.07726/) (2018)
- [From Denoising to Compressed Sensing](https://mlsanity.com/paper/1406.4175/) (2014)
- [Complex Transformer: A Framework for Modeling Complex-Valued Sequence](https://mlsanity.com/paper/1910.10202/) (2019)
- [Building Blocks for a Complex-Valued Transformer Architecture](https://mlsanity.com/paper/2306.09827/) (2023)
- [Dual-Scale Transformer for Large-Scale Single-Pixel Imaging](https://mlsanity.com/paper/2404.05001/) (2024)

## Abstract (arXiv)

Computer-generated holography (CGH) has gained wide attention with deep learning-based algorithms. However, due to its nonlinear and ill-posed nature, challenges remain in achieving accurate and stable reconstruction. Specifically, (i) the widely used end-to-end networks treat the reconstruction model as a black box, ignoring underlying physical relationships, which reduces interpretability and flexibility. (ii) CNN-based CGH algorithms have limited receptive fields, hindering their ability to capture long-range dependencies and global context. (iii) Angular spectrum method (ASM)-based models are constrained to finite near-fields.In this paper, we propose a Deep Unfolding Network (DUN) that decomposes gradient descent into two modules: an adaptive bandwidth-preserving model (ABPM) and a phase-domain complex-valued denoiser (PCD), providing more flexibility. ABPM allows for wider working distances compared to ASM-based methods. At the same time, PCD leverages its complex-valued deformable self-attention module to capture global features and enhance performance, achieving a PSNR over 35 dB. Experiments on simulated and real data show state-of-the-art results.