End-to-End QGAN-Based Image Synthesis via Neural Noise Encoding and Intensity Calibration
This work addresses a specific bottleneck in quantum machine learning for image synthesis, offering an incremental improvement over existing QGAN methods by enabling end-to-end generation.
The paper tackled the problem of direct full-image synthesis using Quantum Generative Adversarial Networks (QGANs) by proposing ReQGAN, which overcomes bottlenecks in noise encoding and intensity calibration, achieving stable training and effective image generation on MNIST and Fashion-MNIST datasets under limited qubit budgets.
Quantum Generative Adversarial Networks (QGANs) offer a promising path for learning data distributions on near-term quantum devices. However, existing QGANs for image synthesis avoid direct full-image generation, relying on classical post-processing or patch-based methods. These approaches dilute the quantum generator's role and struggle to capture global image semantics. To address this, we propose ReQGAN, an end-to-end framework that synthesizes an entire N=2^D-pixel image using a single D-qubit quantum circuit. ReQGAN overcomes two fundamental bottlenecks hindering direct pixel generation: (1) the rigid classical-to-quantum noise interface and (2) the output mismatch between normalized quantum statistics and the desired pixel-intensity space. We introduce a learnable Neural Noise Encoder for adaptive state preparation and a differentiable Intensity Calibration module to map measurements to a stable, visually meaningful pixel domain. Experiments on MNIST and Fashion-MNIST demonstrate that ReQGAN achieves stable training and effective image synthesis under stringent qubit budgets, with ablation studies verifying the contribution of each component.