Generative models are transforming earth observation, enabling high-quality image generation for applications like data augmentation, cloud removal or image inpainting. This study investigates quantum latent diffusion models, integrating quantum variational circuits within latent diffusion processes. Using three ansatzes Rx-Rz-Rx, universal, and matchgate, we compare those models with classical latent diffusion models and generative adversarial networks. The quantum latent diffusion models with the universal ansatz achieved a 21.5% improvement in Fréchet inception distance and a 29.9% improvement in kernel inception distance compared to the best classical counterpart, while maintaining competitive diversity with an inception score of 1.3152. These results demonstrate the efficiency of quantum-enhanced generative models in the specific case of data augmentation for earth observation applications, expecting similar improvements also in other related earth observation tasks.
Generative models are transforming earth observation, enabling high-quality image generation for applications like data augmentation, cloud removal or image inpainting. This study investigates quantum latent diffusion models, integrating quantum variational circuits within latent diffusion processes. Using three ansatzes Rx-Rz-Rx, universal, and matchgate, we compare those models with classical latent diffusion models and generative adversarial networks. The quantum latent diffusion models with the universal ansatz achieved a 21.5% improvement in Fréchet inception distance and a 29.9% improvement in kernel inception distance compared to the best classical counterpart, while maintaining competitive diversity with an inception score of 1.3152. These results demonstrate the efficiency of quantum-enhanced generative models in the specific case of data augmentation for earth observation applications, expecting similar improvements also in other related earth observation tasks.
Quantum diffusion models are rapidly advancing generative remote-sensing analytics by harnessing quantum computation to deliver higher-quality, more realistic satellite images, while converging faster and with substantially fewer trainable parameters than comparable classical diffusers. In this context, we first realise a fully-quantum latent diffusion model (QLDM) whose denoiser is a variational-quantum circuit acting on a 10-dimensionaI latent code. On EuroSAT, focusing on three land-cover classes Forest, Herbaceous Vegetation and SeaLake, QLDM lowers the Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 21.5 % and 29.9 %, respectively, relative to a parameter-matched classical latent diffuser. Yet the severe spatial compression intrinsic to latent processing limits fine-grained detail. To restore image fidelity, we develop a hybrid quantum—classical architecture—the Quanvolutional Conditioned U-Net (QCU-Net)—which inserts entangling quantum layers both at the U-Net bottleneck and as a quanvolutional filter early in the encoder. Trained on the full ten-class EuroSAT RGB set, QCU-Net reaches an FID of 2.57 and a KID of 0.0008, representing 64 % and 76 % improvements over the best classical diffusion baseline (FID = 7.22; KID = 0.0034), and boosts class-conditioning accuracy to 81.7 % versus 62.2 % for the classical model. Thèse gains confirm that embedding quantum circuits within the feature-extraction pipeline yields richer spatial— spectral representations than latent-space quantum processing alone. Our results outline a clear progression in quantum generative modelling for Earth Observation: by first demonstrating with the QLDM that fully quantum denoising is not only feasible but also highly parameter- efficient, then showing how the hybrid QCU-Net architecture restores perceptual fidelity by embedding quantum feature extractors within a classical U-Net backbone, and finally pointing toward a fully pixel-space quantum diffusion model as future work—one that would leverage deep quantum circuits to perform both the forward and reverse diffusion directly in Hilbert space, remove classical auto-encoding steps, and promise even greater image quality and efficiency on fault-tolerant hardware.
The rapid adoption of diffusion models (DMs) in the Earth Observation (EO) domain has unlocked new generative capabilities aimed at producing new samples, whose statistical properties closely match real imagery, for tasks such as synthesizing missing data, augmenting scarce labeled datasets, and improving image reconstruction. This is particularly relevant in EO, where labeled data are often costly to obtain and limited in availability. However, classical DMs still face significant computational limitations, requiring hundreds to thousands of inference steps, as well as difficulties in capturing the intricate spatial and spectral correlations characteristic of EO data. Recent research in Quantum Machine Learning (QML), including initial attempts of Quantum Generative Models, offers a fundamentally different approach to overcome these challenges. Motivated by these considerations, we introduce the Quanvolutional Conditioned U-Net (QCU-Net), a hybrid quantum–classical architecture that applies quantum operations within a conditioned diffusion framework using a novel quanvolutional feature-extraction approach, for generating synthetic labeled EO imagery. Extensive experiments on the EuroSAT RGB dataset demonstrate that our QCU-Net achieves superior results. Notably, it reduces the Fréchet Inception Distance by 64%, lowers the Kernel Inception Distance by 76%, and yields higher semantic accuracy. Ablation studies further reveal that strategically positioning quantum layers and employing entangling variational circuits enhance model performance and convergence. This work represents the first successful adaptation of class-conditioned quantum diffusion modeling in the EO domain, paving the way for quantum-enhanced remote sensing imagery synthesis.
The rapid adoption of diffusion models (DMs) in the Earth Observation (EO) domain has unlocked new generative capabilities aimed at producing new samples, whose statistical properties closely match real imagery, for tasks such as synthesizing missing data, augmenting scarce labeled datasets, and improving image reconstruction. This is particularly relevant in EO, where labeled data are often costly to obtain and limited in availability. However, classical DMs still face significant computational limitations, requiring hundreds to thousands of inference steps, as well as difficulties in capturing the intricate spatial and spectral correlations characteristic of EO data. Recent research in Quantum Machine Learning (QML), including initial attempts of Quantum Generative Models, offers a fundamentally different approach to overcome these challenges. Motivated by these considerations, we introduce the Quanvolutional Conditioned U-Net (QCU-Net), a hybrid quantum--classical architecture that applies quantum operations within a conditioned diffusion framework using a novel quanvolutional feature-extraction approach, for generating synthetic labeled EO imagery. Extensive experiments on the EuroSAT RGB dataset demonstrate that our QCU-Net achieves superior results. Notably, it reduces the Fréchet Inception Distance by 64%, lowers the Kernel Inception Distance by 76%, and yields higher semantic accuracy. Ablation studies further reveal that strategically positioning quantum layers and employing entangling variational circuits enhance model performance and convergence. This work represents the first successful adaptation of class-conditioned quantum diffusion modeling in the EO domain, paving the way for quantum-enhanced remote sensing imagery synthesis.
Generative models have gained attention for their proven effectiveness in various domains, from generating images to producing text. Diffusion models (DMs) have demonstrated superior performance compared to alternative techniques like generative adversarial networks (GANs) or autoencoders (AE). The research about DMs has advanced further with the proposition of latent diffusion models (LDMs), which employ DMs within the latent space to mitigate computational complexity while preserving high quality and flexibility. Quantum computing with its intrinsic characteristics of entanglement and superposition which enable high representational power, arises with the possibility of offering computational advantages over classical generative models. Inspired by current trends of classical generative models, we propose quantum latent diffusion models (QLDMs) to enable computationally efficient DMs with the guidance of conditions on classical data. QLDMs offer solutions to the challenges faced by classical generative models due to their large number of parameters, while still achieving comparable results on datasets.
Quantum machine learning (QML) has attracted increasing attention due to its excellent performance on clas-sification and identification tasks. In quantum machine learning methods, quantum generative adversarial networks (QGANs) are an extension of classical generative adversarial networks in the field of quantum computing. However, current quan-tum generation techniques are limited by quantum computing resources, these techniques are only validated on small-scale datasets, and the images generated by QGANs are of lower quality. To improve the image quality generated by QGANs, we introduce a new method to construct latent spaces based on quantum principal component analysis, to ensure high-quality data output of QGANs. In addition, we employ a parameterized quantum circuit instead of a sub module of quantum principal component analysis (Quantum Phase Estimation) to reduce the circuit depths and the number of gates. Experimental results indicate that our new method, which utilizes the new latent space as the input for QGANs, outperforms the original QGAN in terms of performance metrics Frechet Inception Distance and Cosine similarity. Specifically, our method achieves a 20 % lower Frechet Inception Distance score and 9 % higher cosine similarity compared to the original QGAN.
Generative learning models in medical research are crucial in developing training data for deep learning models and advancing diagnostic tools, but the problem of high quality, diverse images is an open topic of research. Quantum-enhanced generative models have been proposed and tested in the literature but have been restricted to small problems below the scale of industry relevance. In this paper, we propose quantum-enhanced diffusion and variational autoencoder (VAE) models and test them on the Fundus retinal image generation task. In our numerical experiments, the images generated using quantum-enhanced models are of higher quality and match more closely in features to the real image distribution compared to the ones generated using classical diffusion models, even when the classical diffusion models are larger than the quantum model. Additionally, we perform noisy testing and experimental runs on hardware to confirm the numerical experiments, finding that quantum-enhanced diffusion model can sometimes produce higher quality images, both in terms of diversity and fidelity, when tested with quantum noise.