High-spatial-resolution hyperspectral images (HSI) are essential for applications such as remote sensing and medical imaging, yet HSI sensors inherently trade spatial detail for spectral richness. Fusing high-spatial-resolution multispectral images (HR-MSI) with low-spatial-resolution hyperspectral images (LR-HSI) is a promising route to recover fine spatial structures without sacrificing spectral fidelity. Most state-of-the-art methods for HSI-MSI fusion demand point spread function (PSF) calibration or ground truth high resolution HSI (HR-HSI), both of which are impractical to obtain in real world settings. We present SpectraLift, a fully self-supervised framework that fuses LR-HSI and HR-MSI inputs using only the MSI's Spectral Response Function (SRF). SpectraLift trains a lightweight per-pixel multi-layer perceptron (MLP) network using ($i$)~a synthetic low-spatial-resolution multispectral image (LR-MSI) obtained by applying the SRF to the LR-HSI as input, ($ii$)~the LR-HSI as the output, and ($iii$)~an $\ell_1$ spectral reconstruction loss between the estimated and true LR-HSI as the optimization objective. At inference, SpectraLift uses the trained network to map the HR-MSI pixel-wise into a HR-HSI estimate. SpectraLift converges in minutes, is agnostic to spatial blur and resolution, and outperforms state-of-the-art methods on PSNR, SAM, SSIM, and RMSE benchmarks.
Healthy tissue inpainting has significant applications, including the generation of pseudo-healthy baselines for tumor growth models and the facilitation of image registration. In previous editions of the BraTS Local Synthesis of Healthy Brain Tissue via Inpainting Challenge, denoising diffusion probabilistic models (DDPMs) demonstrated qualitatively convincing results but suffered from low sampling speed. To mitigate this limitation, we adapted a 2D image generation approach, combining DDPMs with generative adversarial networks (GANs) and employing a variance-preserving noise schedule, for the task of 3D inpainting. Our experiments showed that the variance-preserving noise schedule and the selected reconstruction losses can be effectively utilized for high-quality 3D inpainting in a few time steps without requiring adversarial training. We applied our findings to a different architecture, a 3D wavelet diffusion model (WDM3D) that does not include a GAN component. The resulting model, denoted as fastWDM3D, obtained a SSIM of 0.8571, a MSE of 0.0079, and a PSNR of 22.26 on the BraTS inpainting test set. Remarkably, it achieved these scores using only two time steps, completing the 3D inpainting process in 1.81 s per image. When compared to other DDPMs used for healthy brain tissue inpainting, our model is up to 800 x faster while still achieving superior performance metrics. Our proposed method, fastWDM3D, represents a promising approach for fast and accurate healthy tissue inpainting. Our code is available at https://github.com/AliciaDurrer/fastWDM3D.
Computed Tomography (CT) is a widely utilized imaging modality in clinical settings. Using densely acquired rotational X-ray arrays, CT can capture 3D spatial features. However, it is confronted with challenged such as significant time consumption and high radiation exposure. CT reconstruction methods based on sparse-view X-ray images have garnered substantial attention from researchers as they present a means to mitigate costs and risks. In recent years, diffusion models, particularly the Latent Diffusion Model (LDM), have demonstrated promising potential in the domain of 3D CT reconstruction. Nonetheless, due to the substantial differences between the 2D latent representation of X-ray modalities and the 3D latent representation of CT modalities, the vanilla LDM is incapable of achieving effective alignment within the latent space. To address this issue, we propose the Consistent Latent Space Diffusion Model (CLS-DM), which incorporates cross-modal feature contrastive learning to efficiently extract latent 3D information from 2D X-ray images and achieve latent space alignment between modalities. Experimental results indicate that CLS-DM outperforms classical and state-of-the-art generative models in terms of standard voxel-level metrics (PSNR, SSIM) on the LIDC-IDRI and CTSpine1K datasets. This methodology not only aids in enhancing the effectiveness and economic viability of sparse X-ray reconstructed CT but can also be generalized to other cross-modal transformation tasks, such as text-to-image synthesis. We have made our code publicly available at https://anonymous.4open.science/r/CLS-DM-50D6/ to facilitate further research and applications in other domains.
The escalating adoption of high-resolution, large-field-of-view imagery amplifies the need for efficient compression methodologies. Conventional techniques frequently fail to preserve critical image details, while data-driven approaches exhibit limited generalizability. Implicit Neural Representations (INRs) present a promising alternative by learning continuous mappings from spatial coordinates to pixel intensities for individual images, thereby storing network weights rather than raw pixels and avoiding the generalization problem. However, INR-based compression of large images faces challenges including slow compression speed and suboptimal compression ratios. To address these limitations, we introduce COLI (Compressor for Large Images), a novel framework leveraging Neural Representations for Videos (NeRV). First, recognizing that INR-based compression constitutes a training process, we accelerate its convergence through a pretraining-finetuning paradigm, mixed-precision training, and reformulation of the sequential loss into a parallelizable objective. Second, capitalizing on INRs' transformation of image storage constraints into weight storage, we implement Hyper-Compression, a novel post-training technique to substantially enhance compression ratios while maintaining minimal output distortion. Evaluations across two medical imaging datasets demonstrate that COLI consistently achieves competitive or superior PSNR and SSIM metrics at significantly reduced bits per pixel (bpp), while accelerating NeRV training by up to 4 times.
Image fusion aims to integrate complementary information across modalities to generate high-quality fused images, thereby enhancing the performance of high-level vision tasks. While global spatial modeling mechanisms show promising results, constructing long-range feature dependencies in the spatial domain incurs substantial computational costs. Additionally, the absence of ground-truth exacerbates the difficulty of capturing complementary features effectively. To tackle these challenges, we propose a Residual Prior-driven Frequency-aware Network, termed as RPFNet. Specifically, RPFNet employs a dual-branch feature extraction framework: the Residual Prior Module (RPM) extracts modality-specific difference information from residual maps, thereby providing complementary priors for fusion; the Frequency Domain Fusion Module (FDFM) achieves efficient global feature modeling and integration through frequency-domain convolution. Additionally, the Cross Promotion Module (CPM) enhances the synergistic perception of local details and global structures through bidirectional feature interaction. During training, we incorporate an auxiliary decoder and saliency structure loss to strengthen the model's sensitivity to modality-specific differences. Furthermore, a combination of adaptive weight-based frequency contrastive loss and SSIM loss effectively constrains the solution space, facilitating the joint capture of local details and global features while ensuring the retention of complementary information. Extensive experiments validate the fusion performance of RPFNet, which effectively integrates discriminative features, enhances texture details and salient objects, and can effectively facilitate the deployment of the high-level vision task.
Magnetic Resonance Fingerprinting (MRF) enables fast quantitative imaging by matching signal evolutions to a predefined dictionary. However, conventional dictionary matching suffers from exponential growth in computational cost and memory usage as the number of parameters increases, limiting its scalability to multi-parametric mapping. To address this, recent work has explored deep learning-based approaches as alternatives to DM. We propose GAST-Mamba, an end-to-end framework that combines a dual Mamba-based encoder with a Gate-Aware Spatial-Temporal (GAST) processor. Built on structured state-space models, our architecture efficiently captures long-range spatial dependencies with linear complexity. On 5 times accelerated simulated MRF data (200 frames), GAST-Mamba achieved a T1 PSNR of 33.12~dB, outperforming SCQ (31.69~dB). For T2 mapping, it reached a PSNR of 30.62~dB and SSIM of 0.9124. In vivo experiments further demonstrated improved anatomical detail and reduced artifacts. Ablation studies confirmed that each component contributes to performance, with the GAST module being particularly important under strong undersampling. These results demonstrate the effectiveness of GAST-Mamba for accurate and robust reconstruction from highly undersampled MRF acquisitions, offering a scalable alternative to traditional DM-based methods.
3D Gaussian Splatting (3DGS) has emerged as a powerful technique for radiance field rendering, but it typically requires millions of redundant Gaussian primitives, overwhelming memory and rendering budgets. Existing compaction approaches address this by pruning Gaussians based on heuristic importance scores, without global fidelity guarantee. To bridge this gap, we propose a novel optimal transport perspective that casts 3DGS compaction as global Gaussian mixture reduction. Specifically, we first minimize the composite transport divergence over a KD-tree partition to produce a compact geometric representation, and then decouple appearance from geometry by fine-tuning color and opacity attributes with far fewer Gaussian primitives. Experiments on benchmark datasets show that our method (i) yields negligible loss in rendering quality (PSNR, SSIM, LPIPS) compared to vanilla 3DGS with only 10% Gaussians; and (ii) consistently outperforms state-of-the-art 3DGS compaction techniques. Notably, our method is applicable to any stage of vanilla or accelerated 3DGS pipelines, providing an efficient and agnostic pathway to lightweight neural rendering.
Super-resolution (SR) is an ill-posed inverse problem with many feasible solutions consistent with a given low-resolution image. On one hand, regressive SR models aim to balance fidelity and perceptual quality to yield a single solution, but this trade-off often introduces artifacts that create ambiguity in information-critical applications such as recognizing digits or letters. On the other hand, diffusion models generate a diverse set of SR images, but selecting the most trustworthy solution from this set remains a challenge. This paper introduces a robust, automated framework for identifying the most trustworthy SR sample from a diffusion-generated set by leveraging the semantic reasoning capabilities of vision-language models (VLMs). Specifically, VLMs such as BLIP-2, GPT-4o, and their variants are prompted with structured queries to assess semantic correctness, visual quality, and artifact presence. The top-ranked SR candidates are then ensembled to yield a single trustworthy output in a cost-effective manner. To rigorously assess the validity of VLM-selected samples, we propose a novel Trustworthiness Score (TWS) a hybrid metric that quantifies SR reliability based on three complementary components: semantic similarity via CLIP embeddings, structural integrity using SSIM on edge maps, and artifact sensitivity through multi-level wavelet decomposition. We empirically show that TWS correlates strongly with human preference in both ambiguous and natural images, and that VLM-guided selections consistently yield high TWS values. Compared to conventional metrics like PSNR, LPIPS, which fail to reflect information fidelity, our approach offers a principled, scalable, and generalizable solution for navigating the uncertainty of the diffusion SR space. By aligning outputs with human expectations and semantic correctness, this work sets a new benchmark for trustworthiness in generative SR.
Accurately identifying oocytes that progress to the blastocyst stage is crucial in reproductive medicine, but the limited availability of annotated high-quality embryo images presents challenges for developing automated diagnostic tools. To address this, we propose BlastDiffusion, a generative model based on Latent Diffusion Models (LDMs) that synthesizes realistic oocyte images conditioned on developmental outcomes. Our approach utilizes a pretrained Variational Autoencoder (VAE) for latent space representation, combined with a diffusion process to generate images that distinguish between oocytes that reach the blastocyst stage and those that do not. When compared to Blastocyst-GAN, a GAN-based model we trained for this task, BlastDiffusion achieves superior performance, with a global Frechet Inception Distance (FID) of 94.32, significantly better than Blastocyst-GAN's FID of 232.73. Additionally, our model shows improvements in perceptual (LPIPS) and structural (SSIM) similarity to real oocyte images. Qualitative analysis further demonstrates that BlastDiffusion captures key morphological differences linked to developmental outcomes. These results highlight the potential of diffusion models in reproductive medicine, offering an effective tool for data augmentation and automated embryo assessment.
The overexpression of the human epidermal growth factor receptor 2 (HER2) in breast cells is a key driver of HER2-positive breast cancer, a highly aggressive subtype requiring precise diagnosis and targeted therapy. Immunohistochemistry (IHC) is the standard technique for HER2 assessment but is costly, labor-intensive, and highly dependent on antibody selection. In contrast, hematoxylin and eosin (H&E) staining, a routine histopathological procedure, offers broader accessibility but lacks HER2 specificity. This study proposes an advanced deep learning-based image translation framework to generate highfidelity IHC images from H&E-stained tissue samples, enabling cost-effective and scalable HER2 assessment. By modifying the loss function of pyramid pix2pix, we mitigate mode collapse, a fundamental limitation in generative adversarial networks (GANs), and introduce a novel variance-based penalty that enforces structural diversity in generated images. Our model particularly excels in translating HER2-positive (IHC 3+) images, which have remained challenging for existing methods due to their complex morphological variations. Extensive evaluations on the BCI histopathological dataset demonstrate that our model surpasses state-of-the-art methods in terms of peak signal-tonoise ratio (PSNR), structural similarity index (SSIM), and Frechet Inception Distance (FID), particularly in accurately translating HER2-positive (IHC 3+) images. Beyond medical imaging, our model exhibits superior performance in general image-to-image translation tasks, showcasing its potential across multiple domains. This work marks a significant step toward AI-driven precision oncology, offering a reliable and efficient alternative to traditional HER2 diagnostics.
Cloud contamination significantly impairs the usability of optical satellite imagery, affecting critical applications such as environmental monitoring, disaster response, and land-use analysis. This research presents a Cloud-Attentive Reconstruction Framework that integrates SAR-optical feature fusion with deep learning-based image reconstruction to generate cloud-free optical imagery. The proposed framework employs an attention-driven feature fusion mechanism to align complementary structural information from Synthetic Aperture Radar (SAR) with spectral characteristics from optical data. Furthermore, a cloud-aware model update strategy introduces adaptive loss weighting to prioritize cloud-occluded regions, enhancing reconstruction accuracy. Experimental results demonstrate that the proposed method outperforms existing approaches, achieving a PSNR of 31.01 dB, SSIM of 0.918, and MAE of 0.017. These outcomes highlight the framework's effectiveness in producing high-fidelity, spatially and spectrally consistent cloud-free optical images.
Cone-beam computed tomography (CBCT) using only a few X-ray projection views enables faster scans with lower radiation dose, but the resulting severe under-sampling causes strong artifacts and poor spatial coverage. We address these challenges in a unified framework. First, we replace conventional UNet/ResNet encoders with TransUNet, a hybrid CNN-Transformer model. Convolutional layers capture local details, while self-attention layers enhance global context. We adapt TransUNet to CBCT by combining multi-scale features, querying view-specific features per 3D point, and adding a lightweight attenuation-prediction head. This yields Trans-CBCT, which surpasses prior baselines by 1.17 dB PSNR and 0.0163 SSIM on the LUNA16 dataset with six views. Second, we introduce a neighbor-aware Point Transformer to enforce volumetric coherence. This module uses 3D positional encoding and attention over k-nearest neighbors to improve spatial consistency. The resulting model, Trans$^2$-CBCT, provides an additional gain of 0.63 dB PSNR and 0.0117 SSIM. Experiments on LUNA16 and ToothFairy show consistent gains from six to ten views, validating the effectiveness of combining CNN-Transformer features with point-based geometry reasoning for sparse-view CBCT reconstruction.
Optical analog circuits have attracted attention as promising alternatives to traditional electronic circuits for signal processing tasks due to their potential for low-latency and low-power computations. However, implementing iterative algorithms on such circuits presents challenges, particularly due to the difficulty of performing division operations involving dynamically changing variables and the additive noise introduced by optical amplifiers. In this study, we investigate the feasibility of implementing image restoration algorithms using total variation regularization on optical analog circuits. Specifically, we design the circuit structures for the image restoration with widely used alternating direction method of multipliers (ADMM) and primal dual splitting (PDS). Our design avoids division operations involving dynamic variables and incorporate the impact of additive noise introduced by optical amplifiers. Simulation results show that the effective denoising can be achieved in terms of peak signal to noise ratio (PSNR) and structural similarity index measure (SSIM) even when the circuit noise at the amplifiers is taken into account.
AI requires extensive datasets, while medical data is subject to high data protection. Anonymization is essential, but poses a challenge for some regions, such as the head, as identifying structures overlap with regions of clinical interest. Synthetic data offers a potential solution, but studies often lack rigorous evaluation of realism and utility. Therefore, we investigate to what extent synthetic data can replace real data in segmentation tasks. We employed head and neck cancer CT scans and brain glioma MRI scans from two large datasets. Synthetic data were generated using generative adversarial networks and diffusion models. We evaluated the quality of the synthetic data using MAE, MS-SSIM, Radiomics and a Visual Turing Test (VTT) performed by 5 radiologists and their usefulness in segmentation tasks using DSC. Radiomics indicates high fidelity of synthetic MRIs, but fall short in producing highly realistic CT tissue, with correlation coefficient of 0.8784 and 0.5461 for MRI and CT tumors, respectively. DSC results indicate limited utility of synthetic data: tumor segmentation achieved DSC=0.064 on CT and 0.834 on MRI, while bone segmentation a mean DSC=0.841. Relation between DSC and correlation is observed, but is limited by the complexity of the task. VTT results show synthetic CTs' utility, but with limited educational applications. Synthetic data can be used independently for the segmentation task, although limited by the complexity of the structures to segment. Advancing generative models to better tolerate heterogeneous inputs and learn subtle details is essential for enhancing their realism and expanding their application potential.
Structural coloration is commonly modeled using wave optics for reliable and photorealistic rendering of natural, quasi-periodic and complex nanostructures. Such models often rely on dense, preliminary or preprocessed data to accurately capture the nuanced variations in diffractive surface reflectances. This heavy data dependency warrants implicit neural representation which has not been addressed comprehensively in the current literature. In this paper, we present a multi-layer perceptron (MLP) based method for data-driven rendering of diffractive surfaces with high accuracy and efficiency. We primarily approach this problem from a data compression perspective to devise a nuanced training and modeling method which is attuned to the domain and range characteristics of diffractive reflectance datasets. Importantly, our approach avoids over-fitting and has robust resampling behavior. Using Peak-Signal-to-Noise (PSNR), Structural Similarity Index Measure (SSIM) and a flipping difference evaluator (FLIP) as evaluation metrics, we demonstrate the high-quality reconstruction of the ground-truth. In comparison to a recent state-of-the-art offline, wave-optical, forward modeling approach, our method reproduces subjectively similar results with significant performance gains. We reduce the memory footprint of the raw datasets by two orders of magnitude in general. Lastly, we depict the working of our method with actual surface renderings.
Cone-beam CT (CBCT) is widely used in clinical radiotherapy for image-guided treatment, improving setup accuracy, adaptive planning, and motion management. However, slow gantry rotation limits performance by introducing motion artifacts, blurring, and increased dose. This work aims to develop a clinically feasible method for reconstructing high-quality CBCT volumes from consecutive limited-angle acquisitions, addressing imaging challenges in time- or dose-constrained settings. We propose a limited-angle (LA) geometry-integrated cycle-domain (LA-GICD) framework for CBCT reconstruction, comprising two denoising diffusion probabilistic models (DDPMs) connected via analytic cone-beam forward and back projectors. A Projection-DDPM completes missing projections, followed by back-projection, and an Image-DDPM refines the volume. This dual-domain design leverages complementary priors from projection and image spaces to achieve high-quality reconstructions from limited-angle (<= 90 degrees) scans. Performance was evaluated against full-angle reconstruction. Four board-certified medical physicists conducted assessments. A total of 78 planning CTs in common CBCT geometries were used for training and evaluation. The method achieved a mean absolute error of 35.5 HU, SSIM of 0.84, and PSNR of 29.8 dB, with visibly reduced artifacts and improved soft-tissue clarity. LA-GICD's geometry-aware dual-domain learning, embedded in analytic forward/backward operators, enabled artifact-free, high-contrast reconstructions from a single 90-degree scan, reducing acquisition time and dose four-fold. LA-GICD improves limited-angle CBCT reconstruction with strong data fidelity and anatomical realism. It offers a practical solution for short-arc acquisitions, enhancing CBCT use in radiotherapy by providing clinically applicable images with reduced scan time and dose for more accurate, personalized treatments.
Semantic communication has emerged as a promising neural network-based system design for 6G networks. Task-oriented semantic communication is a novel paradigm whose core goal is to efficiently complete specific tasks by transmitting semantic information, optimizing communication efficiency and task performance. The key challenge lies in preserving privacy while maintaining task accuracy, as this scenario is susceptible to model inversion attacks. In such attacks, adversaries can restore or even reconstruct input data by analyzing and processing model outputs, owing to the neural network-based nature of the systems. In addition, traditional systems use image quality indicators (such as PSNR or SSIM) to assess attack severity, which may be inadequate for task-oriented semantic communication, since visual differences do not necessarily ensure semantic divergence. In this paper, we propose a diffusion-based semantic communication framework, named DiffSem, that optimizes semantic information reconstruction through a diffusion mechanism with self-referential label embedding to significantly improve task performance. Our model also compensates channel noise and adopt semantic information distortion to ensure the robustness of the system in various signal-to-noise ratio environments. To evaluate the attacker's effectiveness, we propose a new metric that better quantifies the semantic fidelity of estimations from the adversary. Experimental results based on this criterion show that on the MNIST dataset, DiffSem improves the classification accuracy by 10.03%, and maintain stable performance under dynamic channels. Our results further demonstrate that significant deviation exists between traditional image quality indicators and the leakage of task-relevant semantic information.
Image restoration faces challenges including ineffective feature fusion, computational bottlenecks and inefficient diffusion processes. To address these, we propose DiffRWKVIR, a novel framework unifying Test-Time Training (TTT) with efficient diffusion. Our approach introduces three key innovations: (1) Omni-Scale 2D State Evolution extends RWKV's location-dependent parameterization to hierarchical multi-directional 2D scanning, enabling global contextual awareness with linear complexity O(L); (2) Chunk-Optimized Flash Processing accelerates intra-chunk parallelism by 3.2x via contiguous chunk processing (O(LCd) complexity), reducing sequential dependencies and computational overhead; (3) Prior-Guided Efficient Diffusion extracts a compact Image Prior Representation (IPR) in only 5-20 steps, proving 45% faster training/inference than DiffIR while solving computational inefficiency in denoising. Evaluated across super-resolution and inpainting benchmarks (Set5, Set14, BSD100, Urban100, Places365), DiffRWKVIR outperforms SwinIR, HAT, and MambaIR/v2 in PSNR, SSIM, LPIPS, and efficiency metrics. Our method establishes a new paradigm for adaptive, high-efficiency image restoration with optimized hardware utilization.
3D Gaussian Splatting (3DGS) has emerged as a preferred choice alongside Neural Radiance Fields (NeRF) in inverse rendering due to its superior rendering speed. Currently, the common approach in 3DGS is to utilize "single-view" mini-batch training, where only one image is processed per iteration, in contrast to NeRF's "multi-view" mini-batch training, which leverages multiple images. We observe that such single-view training can lead to suboptimal optimization due to increased variance in mini-batch stochastic gradients, highlighting the necessity for multi-view training. However, implementing multi-view training in 3DGS poses challenges. Simply rendering multiple images per iteration incurs considerable overhead and may result in suboptimal Gaussian densification due to its reliance on single-view assumptions. To address these issues, we modify the rasterization process to minimize the overhead associated with multi-view training and propose a 3D distance-aware D-SSIM loss and multi-view adaptive density control that better suits multi-view scenarios. Our experiments demonstrate that the proposed methods significantly enhance the performance of 3DGS and its variants, freeing 3DGS from the constraints of single-view training.
Registration of histological and mass spectrometry imaging (MSI) allows for more precise identification of structural changes and chemical interactions in tissue. With histology and MSI having entirely different image formation processes and dimensionalities, registration of the two modalities remains an ongoing challenge. This work proposes a solution that synthesises histological images from MSI, using a pix2pix model, to effectively enable unimodal registration. Preliminary results show promising synthetic histology images with limited artifacts, achieving increases in mutual information (MI) and structural similarity index measures (SSIM) of +0.924 and +0.419, respectively, compared to a baseline U-Net model. Our source code is available on GitHub: https://github.com/kimberley/MIUA2025.