Papers

Molten pool flow significantly influences bead geometry, defect formation, and solidification microstructure in wire-arc directed energy deposition (DED). However, direct flow sensing remains an open challenge due to extreme temperatures, intense arc illumination, and complex flow patterns. Traditional research has relied either on computationally intensive fluid dynamics simulations, which require assumed boundary conditions and lack empirical anchoring, or on vision-based tracer particle tracking, which yields sparse, noisy, and physically unconstrained velocity estimates. This work bridges these two paradigms by reconstructing physics-constrained 2D surface flow fields from high-speed imaging of a GTAW-based wire-arc DED process under distinct flow regimes: quasi-steady convection and periodic droplet impact. The pipeline begins with homography correction to calibrate the oblique camera view, followed by dense optical flow extraction to provide raw velocity observations. For quasi-steady pools, a physics-informed neural network (PINN) reconstructs the surface velocity field by jointly enforcing data fidelity, incompressible Navier-Stokes momentum equations, no-penetration boundary conditions, and an arc-forcing prior through composite loss minimization. For pools subject to periodic droplet impact, a phase-conditioned time-dependent PINN is further developed to resolve the cyclic flow evolution throughout the impact period. This work provides a unified framework that bridges high-speed visual sensing and fluid mechanics, enabling physics-grounded flow diagnostics for wire-arc DED processes.

Endoscopy is widely used in medicine to visualize anatomical structures for diagnosis and treatment evaluation. However, it does not currently allow rapid, quantitative measurement of airway cross-sectional area (CSA), a key metric for diagnosing and managing respiratory pathologies. There is growing interest in AI-enabled quantitative endoscopy using depth estimation, but a fundamental challenge is the need for ground-truth depth data to train algorithms capable of quantifying depth in physical units (e.g., meters). In this study, 60,000 paired virtual endoscopy images and depth maps were created in Blender from 3D airway models built from computed tomography. A U-Net convolutional neural network was trained to predict the depth maps from the endoscopy images. Point clouds were reconstructed from the AI-generated depth maps using the Equisolid fisheye lens model, and airway perimeters were identified by intersecting a plane with the point cloud to estimate CSA. Nasopharyngeal CSA, a key metric in obstructive sleep apnea, was quantified using this AI-enabled approach with a median error of 5.3 % relative to the 3D model ground-truth. This pilot study using synthetic data provides strong evidence for the feasibility of developing AI-enabled quantitative endoscopy for airway CSA measurement.

On-board compression of synthetic aperture radar (SAR) phase history is bandwidth-critical, and block-adaptive quantization (BAQ) remains the operational standard. We test whether a small convolutional autoencoder, with its encoder on the sensor, can compete with BAQ on complex phase-history patches from the AFRL GOTCHA collection. Every method is charged for all transmitted bits, rates are reported in bits per complex sample (b/cs), and detection is scored by one-to-one matching of CA-CFAR detections. The autoencoder (28,656 encoder parameters) loses at every rate. At 16 b/cs it reaches -2.87 dB NMSE, against -35.5 dB for 8-bit BAQ with $\pm 3\sigma$ clipping and -41.0 dB with a tuned clipping range. It also loses to a $16 \times 16$ block Karhunen-Loève transform (KLT), a local linear coder with a tenth of its encoder cost (-5.39 dB). Running the network in a companded Fourier domain helps, but its detection F1 remains bounded at 33%. The evidence points to this model, its normalization, and its objective, not to a fundamental limit of learned coding. Per patch, the data have modest lag-1 coherence ($|\rho| \approx 0.3$) and patch-specific spectral concentration. Two findings concern evaluation itself. First, 97% of CFAR crossings on raw $64 \times 64$ patches are border artifacts of the zero-padded detector. Second, on interior cells BAQ's clipping range decides detection: 8-bit BAQ keeps 69% F1 with tuned clipping but 17% at $\pm 3\sigma$, and at 8 b/cs or less adaptive FFT thresholding preserves more detections than BAQ. We close with an evaluation protocol for learned radar compression.

Vision-language pretraining objectives differ in the spatial granularity of their supervision, yet the implications of this distribution for chest X-ray interpretation remain underexplored. We present a controlled study that isolates the pretraining objective: holding the encoder and pretraining data fixed, we train nine objectives spanning global and local contrastive learning, captioning, and their combinations, and evaluate across five chest X-ray tasks of increasing spatial granularity. We find that (i) pretraining granularity aligns with task granularity at the extremes, with local objectives leading on abnormality detection and global objectives on classification; (ii) local objectives are surprisingly competitive on global-level generation and question answering tasks; (iii) the merits of captioning and contrastive learning reverse across granularity levels; and (iv) among combinations, mixing captioning and contrastive supervision is strongest on classification and in distribution generation, while pairing two captioning objectives generalizes best on zero-shot report generation. These results show that no single objective is universally optimal, and that the interaction of objective type, granularity, and task governs downstream performance.

Audiovisual Speech Recognition (AVSR) is a multimodal approach to speech recognition that incorporates visual information from lip movements to enhance model performance. Despite its advantages, its development remains constrained by the limited availability of labeled audiovisual (AV) datasets. This work explores the use of synthetic visual data as a solution, using an audio-driven talking-head pipeline to generate lip-synchronized visual content from existing audio data. We evaluate the effectiveness of synthetic visual data both as an augmentation strategy and as a standalone training resource, applying our approach to Spanish and Catalan. Our results show that augmenting real AV data with synthetic samples yields relative Word Error Rate (WER) reductions of up to 16.2%, demonstrating the potential of this approach. Moreover, we demonstrate that synthetic data alone can serve as a baseline for AVSR training in languages lacking AV datasets. These findings provide evidence that synthetic visual data can serve as a scalable solution to AVSR data scarcity, enabling broader language coverage.

Native-latent guidance is a recent paradigm for solving inverse problems with latent diffusion models. It replaces repeated evaluations of the image-space forward model, each requiring a decoder pass, with efficient guidance computed using a learned latent-space surrogate. However, existing methods apply guidance uniformly across latent dimensions, ignoring that measurements are informative only along certain directions and that the reliability of model predictions varies across inputs and timesteps. We propose ANaLOG, a framework for efficient uncertainty-aware guidance with pretrained latent diffusion models. ANaLOG models uncertainty by learning an anisotropic, input- and time-dependent covariance that is integrated into the guidance mechanism to emphasize reliable directions and downweight uncertain ones. We theoretically analyze this framework in a linear model setting and prove that anisotropic, uncertainty-aware weighting is necessary for correct sampling, whereas isotropic guidance induces sampling errors. Experiments across five challenging inverse problems show that ANaLOG improves perceptual reconstruction quality over existing methods while preserving efficiency.

Dormant Awakening (DA) specializes a magnetic resonance imaging (MRI) reconstruction model by fitting a small set of zero weights to one labeled slice. Across fourteen U-shaped convolutional network (U-Net) and vision transformer (ViT) sources, fitting 1,004 or 10,560 weights gives mean peak signal-to-noise ratio (PSNR) gains of .430 and .509 dB. We analyze how scaling the fitted output correction changes evaluation PSNR, then use the calibration correction size to cap changes on new images. The cap bounds changes in root mean squared error (RMSE) relative to the source without an evaluation reference. At 10% adaptation budget, we compare fixed halving and dynamic capping for DA, unrestricted sparse adaptation and low-rank adaptation (LoRA). The controls increase both measured calibration-evaluation correlations across all six tested architecture-adapter settings. Experiments concern one constructed fastMRI-to-M4Raw shift with fixed source checkpoints and participant panels.

Pruning reduces the number of weights in magnetic resonance imaging (MRI) reconstruction networks. Equal sparsity, however, can retain weights with different computational roles and different compatibility with the trained network. We study these effects across 120 convolutional U-Net and vision transformer models using controlled edits evaluated before retraining. In U-Net, preserving high-resolution computation improves reconstruction at equal deletion counts across all tested sparsity levels. Equal-operation controls reveal additional sensitivity of the first convolution, which operation count alone cannot explain. In transformers, retained pretrained weight energy correlates with quality within sparsity levels, yet the same masks reverse their reconstruction ranking when the surrounding trained weights change. At 90% sparsity, matching intermediate output scales largely removes a high-energy-mask penalty while retaining an advantage for the network's original mask. Thus sparsity alone does not characterize reconstruction quality: architecture-specific descriptors are informative, but mask quality can still depend on the surrounding trained network.

In this work, we explore MammoClaw, a training-free agent framework that leverages frozen MLLMs for mammography analysis. To support agentic investigation, we equip the agent with lightweight mammography-specific tools for targeted image analysis, including ROI, paired-view, and contralateral-breast examination. MammoClaw iteratively gathers evidence through these tools, while skill evolution enables non-parametric adaptation by transforming failed trajectories into reusable guidance for later runs. We evaluate the framework on BI-RADS assessment and breast density estimation tasks. In our experiments, we find that tools alone do not reliably improve performance, whereas evolved skills can improve tool-use behavior and performance in some settings. Beyond these results, MammoClaw enables transparent inspection of evidence acquisition, tool interactions, and failure modes, facilitating the analysis and auditing of agent behavior. We view this work as an exploratory study of training-free, self-evolving agentic approaches for mammography and hope it provides a concrete starting point for future work on mammography-specific tools and self-evolution mechanisms. We release our code at https://krishnakanthnakka.github.io/mammoclaw.

Congenital heart disease (CHD) is the most common birth defect, yet a large fraction of cases remain undetected on prenatal ultrasound, in part because current artificial-intelligence methods assume that the key diagnostic frames have already been isolated from a study, by a clinician or by a view classifier. We remove that assumption and address CHD screening directly at the level of the whole ultrasound study. We propose a two-stage framework that first learns transferable frame representations by self-supervised masked-autoencoder pre-training on unlabeled fetal ultrasound, then identifies cardiac frames with a disease-robust module and aggregates them with a transformer-based multiple instance learning (MIL) model that produces a case-level diagnosis from study-level labels alone. The model further returns its highest-scoring frames for clinician review, and a hierarchical head separates critical from non-critical CHD. On the internal test set of our multi-source development cohort (FUSE), the proposed cardiac-gated MIL model reaches an area under the curve (AUC) of 0.985 with a specificity of 0.990, outperforming the reproduced NATMED ensemble (AUC 0.861, specificity 0.600) and the FetalCLIP foundation model (AUC 0.867, specificity 0.710). On an independent external cohort, all models initially perform near chance, but label-free CORAL adaptation raises the proposed model from an AUC of 0.513 to 0.944, whereas whole-study and view-dependent baselines do not recover. These results indicate that whole-study MIL with disease-robust cardiac-frame identification is an accurate and deployable route to prenatal CHD screening.

Large-scale Digital Surface Models (DSMs) can be produced cost-effectively from satellite images via stereo-photogrammetry. However, the resulting 3D maps are often contaminated by noise, outliers, and voids. On the other hand, aerial LiDAR provides high-accuracy elevation measurements at a substantially higher cost. In this work, we study diffusion models conditioned both on photogrammetric DSMs and Pléiades imagery to refine vertically co-registered DSMs. We introduce a modified Stable Diffusion 3 architecture with a pruned text stream and a patch-wise normalization strategy, enabling stable training on LiDAR data and transfer from natural images to elevation maps. Experiments in French cities demonstrate that multimodal conditioning improves elevation accuracy, reducing Dense Urban RMSE from 6.00 to 3.45 m in the in-context cities and from 4.16 to 2.77 m in the held-out city of Bordeaux.

Quantum Machine Learning is a novel field of research aimed at devising machine learning approaches exploiting principles of quantum mechanics, such as superposition, entanglement and interference. In this context, we present a scalable hybrid Quantum Diffusion Model, and evaluate its use for medical image analysis. Specifically, our method is based on a Discrete-Time Quantum Walk algorithm, executed on a real quantum device, to model the forward dynamics of the diffusion model. For the backward step of the diffusion model, we devise and evaluate a classical learning model, which is used to reversely denoise the data. In contrast with other existing attempts at applying quantum machine learning for image analysis tasks, severely limited by the size of existing quantum devices, our method allows to process real-world large size medical data. In particular, we present results on grayscale and RGB images, as well as 3D volumes of moderate sizes. We benchmark our results by reproducing an alternative classical counterpart model, based on diffusion models on discrete state spaces. By doing so, we compare the generation capabilities of both models in terms of three distinct state-of-the-art metrics in the field of image generation, showing the competitive, promising results of our approach.

Polarization imaging captures distinctive surface and geometric cues that benefit a wide range of vision tasks. However, real-world polarization acquisition is often affected by multiple coupled degradations, making image restoration essential for practical polarization vision. Existing methods are largely tailored to specific degradations and remain constrained by the limited scale and quality of polarization data. To address these limitations, we develop an all-in-one polarization restoration framework for diverse and composite degradations. We first study the impact of different polarization representations on restoration performance and identify the normalized Stokes representation as an effective choice for separating intensity and polarization information. Accordingly, we devise a dual-branch architecture that separates intensity and polarization modeling. To overcome the limitations of polarization-specific training, the intensity branch leverages pretrained general restoration priors and a mixture-of-experts extension for composite degradations, while its restoration knowledge is adaptively distilled into the symmetric polarization branch via a cross-domain feature transform. In addition, we establish a composite-degradation polarization benchmark to support all-in-one restoration research. Extensive experiments on public datasets and our proposed benchmark demonstrate the effectiveness of the proposed method.

Iterative algorithms for inverse imaging problems split reconstruction into alternating subproblems, one of which requires evaluating the data-fitting proximal map. In many practical cases, this proximal map has no analytic solution and so must be approximated by an inner loop solver at every iteration, compounding the cost of the outer reconstruction loop. To address this, we propose the pseudo-proximal map (PPM), a reformulation of the data-fitting proximal map as the minimum mean square error estimate of a synthetic probabilistic model. We implement the deep PPM as a self-supervised neural network trained only on sampled Gaussian noise, requiring no ground-truth training images. The deep PPM can be trained for any operator for which the forward model $A$ and its transpose $A^T$ can be evaluated, with provable equivalence to the proximal map when $A$ is linear. We validate the deep PPM on Gaussian deblurring and 4x super-resolution, where the proximal map has an analytic solution, and on X-ray computed tomography (XCT), where no analytic solution exists. Used as the data-fitting agent within an iterative reconstruction method, the deep PPM reproduces the reference reconstruction to within 1% NRMSE for all three operators, and for XCT, it replaces the inner conjugate-gradient loop with a single network evaluation that is 18x faster.

Fourier phase encodes important spatial image structure, but recovering an image without measured spectral magnitude requires additional constraints and leaves absolute intensity ambiguous. We propose a projection-based two-stage framework that combines Fourier-phase and spatial-support constraints with an image-specific neural prior. The first stage alternates constraint enforcement with regularized neural-prior updates, while the second performs phase/support refinement alone with guaranteed convergence. We evaluate two neural-prior implementations on the same 77 microscopy images and compare them with a constraint-only baseline. After 500 final refinement passes, the best-performing variant achieves 31.41 dB pooled PSNR, 35.75 dB mean PSNR, and 0.9531 mean SSIM, improving pooled PSNR by 1.51~dB and reducing pooled MSE by 29.3% relative to the baseline. The results demonstrate the benefit of combining neural guidance with explicit constraint refinement at the evaluated iteration budget, while showing that lower phase residual alone does not guarantee greater reconstruction accuracy.

Magnetic resonance imaging involves an inherent trade-off among spatial resolution, acquisition time, and noise. This trade-off contributes to long scan times and high cost. Deep learning has improved image denoising, but cardiac MRI remains difficult because high-resolution, rapid acquisitions generally lack corresponding low-noise ground truth. Self-supervised denoising offers a potential solution by learning from noisy image pairs or even single noisy acquisitions. However, we show that Noise2Void-style blind-spot denoising, which uses a mean squared error (MSE) loss and assumes zero-mean, independent and identically distributed (i.i.d.) noise, is poorly suited to MR magnitude images. When applied to short-axis $T2$-weighted and $T1\rho$-weighted cardiac MRI with synthetic Rician noise, it produces biased denoised images and biased parametric maps of $T2$ and $T1\rho$. To address this limitation, we formulate self-supervised denoising as maximum likelihood estimation under a known Rician noise model. This yields unbiased denoisers that are competitive with supervised baselines.

Magnetic Resonance Imaging (MRI) often suffers from low signal-to-noise ratio (SNR) and limited spatial resolution, which compromise clinical precision. This study aims to address these challenges by developing a physics-guided degradation model (PGDM) and a novel deep learning framework, Magnetic Resonance Super-Resolution GAN Imaging (MRSRGAN), for improved super-resolution reconstruction of MRI images. The proposed approach consists of two stages: (1) a physics-guided degradation model that simulates low-field MRI conditions by incorporating artifacts such as blur, B0 inhomogeneity, chemical shift effects, down-sampling, and noise to generate paired low-resolution (LR) and high-resolution (HR) training datasets; and (2) an MRSRGAN-based reconstruction network utilizing a hybrid loss function, including modified spectral angle mapper (MSAM) and sharpened ground truth images based on de-convolution of slice profile-based kernels to restore high-SNR and high-resolution images. Experimental validation demonstrates that the proposed method achieves superior spatial resolution, enhanced SNR, improved peak signal-to-noise ratio (PSNR), higher structural similarity index (SSIM), reduced MSAM error, and better non-reference image quality evaluation (NIQE) metrics. Furthermore, it effectively reduces artifacts and demonstrates robustness on real MRI datasets. The MRSRGAN framework, guided by the physics-informed degradation model, provides a significant improvement in MRI image quality, enhancing spatial resolution and diagnostic accuracy. Its demonstrated robustness and effectiveness on real MRI datasets highlight its potential to resolution enhancement for precise diagnostic imaging.

Block-based video codecs select coding parameters based on the input by optimizing a rate-distortion trade-off. The conventional distortion choice, the sum of squared errors (SSE), simplifies parameter selection: the SSE is the sum of block-wise SSEs, so rate-distortion optimization (RDO) can treat blocks independently. Alternatively, full-reference image quality assessment (FR-IQA) metrics such as MS-SSIM or LPIPS often align better with the human visual system than SSE, but they cannot be used in-loop: they do not decompose block-wise and typically require the fully decoded image as input. Building on existing results in metric quadratization, we approximate a broad class of FR-IQA metrics by an input-dependent quadratic distortion (IDQD), whose quadratic form matrix is derived from the Hessian of the metric evaluated at the source video. To make the distortion computable block-wise, we propose two approximations of the Hessian matrix: 1) keeping the block-diagonal, and 2) keeping only its diagonal. We propose estimators for both that require only matrix-vector products with the Hessian obtained by automatic differentiation. Across five metrics for Kodak and CLIC in VVC, IDQD-RDO achieves 14.2-36.7 % BD-rate savings under the target metric with no decoder changes and incurs 10-30 % encoding complexity overhead.

High-resolution digital elevation models (DEMs) support Earth observation applications, but paired training references are often available only at coarser output resolutions. Reconstructing finer terrain grids therefore requires both effective transfer beyond the supervised scale and control of dense-query computation. To address this problem, SCOPE learns a continuous terrain representation from coarser-resolution pairs. It predicts a latent coefficient field on the low-resolution grid and reuses local Fourier residual functions through basis evaluation and geometry-guided ensemble fusion. This separates high-dimensional coefficient prediction from output-grid construction. Experiments on geographically distributed land–ocean samples assess supervised reconstruction, unseen-scale inference, cross-domain generalization, and theoretical computation. SCOPE leads the compared methods across six metrics in the main supervised-scale evaluation. At an unseen factor three times the training factor, land reconstruction reduces RMSE and MAE by approximately 12% relative to bicubic interpolation, with errors close to target-scale fine-tuning. Ninefold output density increases counted multiply–accumulate operations by only about 2%. Frozen-model validation on held-out external marine regions reduces RMSE relative to the DEM-specific implicit baseline EBCF-CDEM by approximately 19% under self-downsampling and 2% with cross-product inputs, while also yielding lower RMSE than LIIF-MS in both settings. These results demonstrate the value of reusable coefficient fields for accurate reconstruction beyond the supervised resolution with low incremental arithmetic cost.

Learned image compression (LIC) is hard to deploy on severely resource-constrained FPGAs, since how fast it actually runs depends not just on arithmetic count, but also on memory traffic, imbalance between different operations, and how the hardware batches its work. We present VQ-LIC, an asymmetric edge-cloud codec in which a compact INT8 depthwise (DW)-pointwise (PW) analysis transform and multi-codebook vector quantization (VQ) run at the edge on a reusable DW/PW engine pair, while reconstruction is handled by a larger cloud decoder. Since VQ codeword matching is expressible as a dot product, it is mapped directly onto the same PW engine, removing the need for a separate VQ compute array, to our knowledge a first for FPGA LIC. A novel latency model, derived from deterministic RTL cycle counts of an FPGA's read, DW, PW, and write costs, predicts an analysis transform's per-block latency; since VQ shares the same PW datapath, the model applies to VQ as well. Validated directly against silicon, the model predicts deployed analysis and VQ latency within 0.26% and 0.05%, and guides the selection of a three-block $16$-$48$-$64$ transform. Post-training codebook reduction then cuts VQ arithmetic and codebook storage by $4\times$ and shrinks the fixed-width latent representation. On a 220-DSP Zynq-7020, VQ-LIC's mid-rate preset reaches 0.1398 bits per pixel at 28.69 dB PSNR and 13.06 dB MS-SSIM on CLIC~2017, outperforming a similarly sized neural encoder and reaching a rate-distortion range comparable to a codec three orders of magnitude larger. The complete 0.1945-kMAC/pixel analysis-VQ pipeline runs at 47.98 frames per second and 42.84 mJ per frame on silicon, using an order of magnitude fewer DSPs than comparable FPGA LIC accelerators while achieving lower bitrate, higher throughput, and lower energy per frame at a modest PSNR tradeoff.

The classification of hyperspectral images (HIs) still presents several challenges. One of them is the difficulty to obtain a large set of labeled samples to train the classifier. Semi-supervised learning methods have received much attention recently, as they require the initial labeling of a reduced number of image pixels and lead to very good results for practical application. One of the open problems in graph-based semi-supervised HI classification is the need to consider relative spatial relationship between pixels in the image to improve the smoothness of the solution. Kernel-based approaches using composite kernels have led to very good results, in which one kernel addresses the spectral properties of the pixels while a second kernel addresses some spatial properties. Most available solutions employ a spectral-spatial kernel which considers the spectral properties of a spatial region about each pixel. This work proposes a composite kernel approach that includes a third kernel dealing exclusively with the relative spatial position of the pixels. Experiments with real HI images show that the use of the new spatial kernel has led to improved classification results when compared to those previously reported in the literature. These results also shed some light on the relative contributions of the spectral and spectral-spatial kernels in semi-supervised HI classification.

Iterative model-based reconstruction in photoacoustic tomography repeatedly applies the forward operator mapping the initial pressure to the transducer signals, and its adjoint. At the scale of current three-dimensional systems, this operator cannot be stored and must be evaluated matrix-free, while accounting for the finite, focused surface of the transducers, whose spatial impulse response degrades the resolution when ignored. Representing the initial pressure by compactly supported radial functions, we show that the measured signal is exactly a temporal convolution between a system kernel gathering the radial function and the electrical impulse response, and a purely geometric quantity accounting for the portion of the transducer surface reached by the wave emitted from a voxel during one time step. Two implementations are proposed, differing only in how this quantity is evaluated: a quadrature over points of the surface, as in existing works, or a closed-form area, which never discretizes the surface. We derive closed forms for planar and cylindrically focused transducers and provide, in the latter case, two accelerations of the resulting elliptic integrals, a lookup table and a trapezoidal approximation, together with the piecewise planar approximation customary in the literature. These implementations reduce the per-voxel geometric computations and are released as an open-source Python package for graphics processing units. The performance of these operators is first demonstrated on a synthetic phantom, where the lookup-table-based operator reaches the accuracy of the exact evaluation ten times faster and outperforms the point discretization on both accuracy and runtime. A second experiment shows that they enable the processing of a realistic vascular phantom at full scale, with a higher peak signalto-noise ratio and a better resolution than the back-projection counterpart. The released implementations are an important step towards the adoption of three-dimensional modelbased photoacoustic reconstructions with finite and focused transducers.

The fidelity of image reconstruction from projections in linear inverse problems, such as tomography, is critically dependent on the synergistic interaction between frequency-domain filtering and spatial-domain interpolation. However, a physical model that can quantitatively describe how these two components cascade interact in the frequency domain and ultimately determine image quality is still lacking to this day. Here, we introduce a unified frequency-domain model that conceptualizes the combined effect of filtering and interpolation in the filtered backprojection (FBP) algorithm as a cascaded modulation process. This model demonstrates that the effective reconstruction spectrum is determined by the original projection data being sequentially modulated by the frequency responses of the filter and the interpolation kernel. Comprehensive numerical simulations and synchrotron radiation CT experiments validate the model, confirming its power to explain the performance hierarchy of classical filter-interpolation pairs under both ideal and noisy conditions. The model successfully predicts key performance characteristics, including spatial resolution and structural fidelity, thereby elucidating the physical principles behind the efficacy of specific combinations. This work establishes a generalizable theoretical foundation for analyzing cascaded systems in linear inverse problems, moving the practice of algorithm selection in computational imaging from empiricism to a principled, physics-based paradigm.

Coordinate-based overfitting image codecs have attracted increasing attention for their low decoding complexity and independence from cross-image generalization. However, representative approaches such as COOL-CHIC face an inherent entropy-modeling trade-off: lightweight models have limited capacity, while more expressive ones incur additional bitrate overhead from transmitting image-specific parameters. Inspired by the prediction mechanism in traditional codecs, we propose a new entropy-modeling strategy that introduces complementary prediction modes with region-adaptive soft mode selection, rather than relying on a single learned predictor to model diverse types of redundancy. Based on this concept, we develop a Long-Range Context Extrapolation Enhanced Entropy Model (LC3EM), which can be integrated into coordinate-based overfitting codecs. Specifically, a parameter-free Neighborhood-based Linear Extrapolation Mode (NLEM) complements the tiny MLP-based local predictor to exploit long-range contextual redundancy and strongly directional structures. A Minimum-Entropy-Inspired Continuous Mode Selection strategy is designed to adaptively fuse these two complementary modes, while requiring the transmission of only the parameters of a single additional linear layer. Moreover, to alleviate the mismatch between training-time relaxed and actual discrete quantization, we introduce a lightweight iterative latent rounding refinement stage to improve compression performance. Experiments demonstrate consistent improvements across diverse benchmarks, particularly on highly regular computer-generated images. When integrated with COOL-CHIC 4.0, the proposed method achieves BD-rate gains of -3.43% and -7.69% on the SIQAD and API datasets, respectively. With COOL-CHIC 5.0 as the backbone, the corresponding gains are -2.88% and -3.15%, respectively. The code will be made publicly available soon.

Diffusion MRI microstructure mapping (MM) is conventionally solved voxel by voxel, ignoring the fact that tissue microstructure forms a spatially organized field. This isolation leaves each estimation problem ill-posed and nonconvex. We instead cast MM as a single global inverse problem, reconstructing the entire parameter field jointly from all measurements of a subject. An untrained neural representation supplies implicit spatial priors and eases the nonconvex optimization, requiring no training data, while coregistered T1-weighted anatomy contributes structural guidance that is freely available in standard protocols. On both synthetic and in-vivo data, our method compares favorably with established voxel-wise and learning-based baselines, suggesting global inversion is a promising alternative.

Video anomaly detection for patrol robots and surveillance systems must recognize abnormal interactions among familiar entities. Existing pixel-reconstruction and isolated-entity methods may fail when individual entities appear normal but their spatial or motion relations are abnormal. This work presents Interaction-Centric Network for Temporal Entity-Relation Analysis and Consistency Testing (INTERACT), a framework for unsupervised video anomaly detection. Person-Object Appearance-Motion Interaction (POAMI) moves beyond isolated-entity encoding by jointly modeling the appearance, motion, and spatial configurations of persons and objects in each frame. The resulting representations capture both individual entity states and their cross-entity context. Based on these representations, Target Geometry-Guided Relational Interaction Prediction (TGRIP) predicts target entity interaction states from historical relational memory and target-frame geometry without explicit identity tracking. Motion-Interaction Reconstruction and Alignment (MIRA) then evaluates these predictions through conditional flow reconstruction and semantic consistency checking, providing complementary anomaly evidence beyond prediction error alone. INTERACT achieves state-of-the-art performance, obtaining a frame-level AUC of 84.5% on the ShanghaiTech benchmark. Ablation studies show that removing cross-entity attention causes the largest performance drop, demonstrating the necessity of relational modeling. INTERACT is particularly effective for anomalies caused by changes in relations among people and objects, while maintaining competitive performance in general scenarios. The source code will be made publicly available at https://github.com/ppworkhard/INTERACT.

Multimodal large language models (MLLMs) must balance local detail against scene context when interpreting ultra-high-resolution (UHR) remote sensing (RS) imagery within a limited visual-input budget. Existing selection-based methods either prune tokens and select patches through relevance scoring, or crop actively through repeated inspection. Neither strategy directly redistributes pixels within a continuous full-scene view: the first retains selected tokens or patches, and the second re-encodes a crop detached from its surroundings. Our pilot study finds that a frozen MLLM already produces useful question-guided spatial requests, yet crop-based inspection of the selected regions does not consistently improve its answers. We therefore formulate UHR understanding as a question of where to spend a fixed pixel budget. Based on this, we introduce GazeEarth, a simple-yet-effective training-free framework that couples question-guided region selection with full-scene foveated observation. The MLLM selects evidence cells from an indexed overview; a deterministic, topology-preserving warp resamples the original image onto a fixed-size canvas, enlarging their shared neighborhood while compressing the periphery; the same frozen model answers from this focused view, using at most two MLLM calls and no external selector or iterative search. Across three UHR remote sensing benchmarks and four frozen backbones, GazeEarth improves benchmark-averaged accuracy by 4.6 to 9.4 percentage points over direct answering and 3.4 to 4.3 over overview answering, outperforming task-trained methods. Our analyses show that existing MLLMs can guide where to look in UHR images on their own, and that what they can infer from the selected evidence depends on how that evidence is presented.

Ultrasound imaging increasingly targets portable, point-of-care, and wearable settings where constraints on power, bandwidth, and hardware complexity often necessitate sparse data acquisition in spatiotemporal scanning. However, image reconstruction using the sparse data can introduce insufficient phase information in coherent beamforming process, resulting in grating-lobe artifacts that degrade imaging contrast resolution. We present a physics-guided, data-driven framework for sparse-to-dense radio-frequency (RF) reconstruction that aligns training with downstream image formation. Our approach trains an end-to-end interpolation network using a hybrid supervision scheme that combines an RF-domain and a beamforming-domain loss with exponential moving average (EMA) to stabilize the multi-objective training. To improve generalization under variable acquisition layouts, we also introduce a random-skip masking strategy that varies sparsity patterns during training so a single model can handle diverse decimation factors and irregular channel configurations. We evaluate the framework on a held-out test set using the mean structural similarity index measure (SSIM) between reconstructed and ground-truth beamformed images. Across decimation factors $\times 2$ to $\times 13$, the best-performing configuration maintains mean SSIM around 0.95. Overall, the results show consistent gains in RF reconstruction and post-beamforming image quality across diverse acquisition conditions. This approach enables robust, high-quality ultrasound imaging at resource-constrained settings by allowing more sparse scanning in spatiotemporal domain.

Primary bone tumors are rare but clinically aggressive neoplasms whose diagnosis from radiographs is challenged by heterogeneous morphology, subtle lesion margins, and overlapping bone structures. To address the limitations of existing single-view models, we present a dual-input, multi-task learning framework that, to our knowledge, is the first to apply bidirectional cross-modal attention between a lesion crop and the full radiograph for joint segmentation and subtype classification. Using the multi-institutional Bone Tumor X-ray Radiograph Dataset (BTXRD, n=3,746), we employ a YOLO-based detector to generate regions of interest, which are paired with full images as inputs to a dual-stream DenseNet121 architecture. Features are integrated via a novel cross-modal attention fusion strategy, refined by Hierarchical Multi-scale Feature Fusion, effectively balancing fine-grained lesion detail with global anatomical context. Evaluated on a held-out patient-level test split, the model demonstrates superior performance over single-input baselines, achieving an overall Dice Similarity Coefficient of 0.896 and a macro-averaged classification F1-score of 0.928. Notably, the system exhibits exceptional sensitivity for malignant osteosarcoma (AUC 0.999), validating the potential of dual-stream context modeling to support radiologists in accurate, early decision-making.

Adolescent Idiopathic Scoliosis is defined as a lateral curvature of the spine that develops during adolescence, without known cause. The condition can result in significant pain and disability, and often progresses rapidly during adolescence. The objective of this paper is to predict the progression of the condition in a temporal sequence from ages 9 to 24, as measured from a sequence of Dual X-ray Absorptiometry (DXA) scans. To this end, we train a transformer model that takes in the curve of the spine to predict curve progression. The model is trained using a large-scale synthetic dataset of spine curves and their time series, covering different curve types and different progression patterns. We show that the model is able to generalise from synthetic to real data by evaluating it on a dataset of real DXA scans covering multiple time points. We find that fine-tuning the model on real data gives a significant boost to performance. The model is able to accurately predict spine curve progression in both scoliosis and normal cases.