Papers

Single-image HDR reconstruction requires inferring missing detail while preserving the visible content of an LDR image. Differences in sensor dynamic range and exposure cause LDR images to lose varying amounts of information in shadows and highlights. We present ExpandDiff, a conditional diffusion pipeline that jointly reconstructs clipped shadows and highlights. To account for this variation, we introduce Dynamic Clipping Synthesis (DCS), which randomly samples shadow and highlight clipping percentiles when constructing training inputs from HDR targets. A pixel-space diffusion model guided by spatially-adaptive normalization then predicts perceptually encoded HDR through a bounded output head, reconstructing both clipping directions in one sampling trajectory. On the SI-HDR benchmark, ExpandDiff variants improve HDR reconstruction accuracy by 3.43 dB in PU21-PSNR over the strongest evaluated competing method, and by 7.34 dB under two-sided clipping. The code and supplementary material are available at https://memreandiran.github.io/expanddiff/.

In knee arthroscopy surgery, accurate registration between preoperative and intraoperative anatomy is a critical step for patient-specific navigation. Achieving an accurate registration requires a reliable 3D reconstruction of the joint during surgery. Preoperative 3D models can be obtained from patient imaging through segmentation and reconstruction, but generating an intraoperative 3D representation remains particularly challenging. Arthroscopic imaging suffers from a limited field of view, low surface texture, and strong specular reflections, which make conventional feature-based 3D reconstruction methods unreliable. In this work, we investigate the application of MIS-NeRF (Minimally-Invasive Surgery Neural Radiance Fields) for reconstructing intraoperative knee 3D models from monocular arthroscopic images. The approach is evaluated on six simulated arthroscopic acquisitions representing six patient-specific knee 3D models. Both qualitative and quantitative results are presented to assess the reconstruction quality. The reconstructed knee 3D models were evaluated through their rendered images, achieving PSNR (Peak Signal-to-Noise Ratio) of 31.88 $\pm$ 2.82, SSIM (Structural Similarity Index) of 0.98 $\pm$ 0.004 and LPIPS (Learned Perceptual Image Patch Similarity) of 0.017 $\pm$ 0.006. These preliminary results suggest the feasibility of NeRF-based reconstruction in the challenging context of arthroscopy and may represent a promising step toward accurate in-silico preoperative-to-intraoperative 3D registration for computer-assisted orthopedic surgery. Further, validation on real arthroscopic data will be necessary to assess clinical applicability.

Text-to-image diffusion models are personalized to a subject by DreamBooth fine-tuning on a handful of its images. Increasingly, these images come from a diffusion model rather than a camera. We show that fine-tuning on such synthetic images degrades subject fidelity, producing oversaturated color and excess high-frequency detail. To isolate the cause, we fine-tune two models from the same base model with the same DreamBooth recipe, one on real photos of a subject and one on synthetic images of that subject generated by the first. We trace the degradation to classifier-free guidance (CFG). For the model personalized on synthetic images, the angle between the conditional and unconditional noise predictions, and with it the norm of their difference, is much larger than for the model personalized on real photos. This inflation grows toward high frequencies and also appears at other prompts semantically close to the subject, such as its class noun, but not at unrelated ones. We propose ReGain, a training-free correction applied at sampling time that measures how much each frequency band of the guidance is inflated relative to the base model and scales that band down accordingly. ReGain needs no real photos. On Stable Diffusion v1.5, ReGain closes 51-64% of the subject-fidelity gap to the model personalized on real photos, as measured by DINO, DINOv2 and CLIP-I. It also improves subject fidelity on SDXL and SD 3.5 and preserves text alignment on all three backbones.

Longitudinal medical imaging analysis is a cornerstone of modern medical practice and patient care. Deep learning applied to longitudinal imaging offers wide potential to enhance diagnosis and track disease progression by capturing spatial changes over time. With major advances in single-timepoint deep learning for imaging, there has been growing interest in longitudinal image analysis, given its increased clinical relevance, though technical challenges remain. Several recent innovations may lead to a new era of multi-timepoint image evaluation, yet the scientific landscape, recent progress, and areas of need remain under-characterized. To address this gap, we conducted a scoping review of deep learning methodologies applied to longitudinal medical imaging, yielding 102 studies published between 2018 and 2025. Neurological disorders (48%) and ophthalmic conditions (12%) were the most common clinical applications, with MRI serving as the predominant imaging modality (67%). Sequential feature modeling approaches combining convolutional neural networks (CNNs) with temporal models (LSTM/RNN) were the most frequent methodology (40%), followed by direct feature aggregation across timepoints (23%). Most studies targeted classification tasks (56%), while external validation was performed in only 24% of studies. Our findings highlight that deep learning-based longitudinal imaging analysis remains a promising field, though newer temporal architectures and larger datasets may improve success and clinical adoption of these tools.

Collaborative perception enables connected agents to share complementary observations for 3D object detection, extending sensing range and mitigating occlusion. Under asynchronous communication, however, cooperative features arrive with temporal delay. Existing prediction-based methods compensate for these features mainly from the transmitting agent's own history, leaving residual misalignment with the ego agent's current observation; subsequent fusion also often overlooks spatial variations in alignment quality. We propose EgoRefine, an ego-referenced predictive alignment and reliability-aware fusion framework for asynchronous collaborative perception. Its Ego-referenced Predictive Alignment module uses the current ego feature to guide cooperative trajectory-field prediction and refines the sampling offsets along an ego-referenced trajectory direction. Its Trajectory-conditioned Reliability-aware Fusion module treats the trajectory discrepancy between the ego and cooperative streams and the directional refinement magnitude as alignment cues, using them to condition the relation between aligned features and adaptively reweight the two streams before convolutional fusion. Experiments on V2V4Real and DAIR-V2X-Seq show that EgoRefine outperforms TraF-Align by 1.6 and 2.9 points on average in AP@0.5 and AP@0.7, respectively. The source code will be made publicly available at https://github.com/godk0509/EgoRefine.

Simplified optical systems often exhibit residual lens aberrations and Veiling Glare (VG), resulting in spatially varying blur and contrast reduction. Large-scale Lens Libraries (LensLib) enable reusable aberration correction models by covering diverse Point Spread Functions (PSFs), but their aberration-only training distribution does not include target-specific veiling glare. Extending such foundations to compound degradation is challenging because realistic target-system compound pairs are difficult to obtain. To address this challenge, we propose LensBridge, a two-stage framework that first establishes a reusable aberration correction foundation and then adapts it to compound optical degradation using only a few unpaired target observations. In Stage I, we build a PSF-aware one-step diffusion foundation by constructing discrete degradation priors from LensLib PSFs and learning to retrieve them directly from aberrated images, enabling PSF-aware correction without requiring explicit PSF at inference. In Stage II, we adapt this foundation to compound degradation through frequency-domain guidance. At the data level, Frequency-guided Degradation Completion (FDC) transfers target low-frequency characteristics to LensLib aberrated images while preserving aberration structures to synthesize compound training pairs; at the model level, Frequency-guided Pseudo Decomposition (FPD) forms aberration- and VG-dominant pseudo observations to condition separate adaptation branches. Extensive experiments across multiple optical systems demonstrate that LensBridge effectively extends reusable aberration correction foundations to joint aberration correction and veiling glare removal without target-system paired supervision. All code will be available at https://github.com/XiaolongQian/LensBridge.

4D flow MRI measures time-resolved, three-directional blood velocity but requires long acquisition times, and its diagnostic signal is carried by the phase difference between velocity encodings, not by image magnitude. Recent work has developed a per-encoding variational network to address image reconstruction in this field. In this work, we incorporate a joint supervised and self-supervised training regime and utilize both magnitude and velocity data during supervision. At the same time, we add multiple acquisition-robust and conditioning strategies based on the acceleration factors. On the CMRx4DFlow~2026 aortic dataset (1.5 and 3T), our model lowers RelErr by $38$–$50\%$ and AngErr by $7.0$–$8.7^\circ$ against a training-matched baseline across $R{=}10$–$50$, improving on every held-out subject at every acceleration. Our model also shows strong generalization ability to transfer on out-of-distribution data by employing the joint training scheme, with an increase of $7.2\%$ in SSIM and decrease of $38\%$ and $31\%$ in AngErr and RelErr respectively.

3D Gaussian Splatting (3DGS) is a popular representation for novel view synthesis. However, 3DGS contains millions of Gaussian primitives, each with rich attributes, resulting in large file sizes. We propose a novel 3DGS compression method based on Teacher-Student Graph Learning (TSGL) that operates directly on a trained model, without 3DGS retraining or access to training images. Specifically, for each block of Gaussian primitives, using decoded positions and DC spherical harmonic (SH) coefficients as predictors, we learn a signal-dependent geometry graph G encoding the pairwise similarities between neighbouring Gaussians via a teacher-student model. Given G, we perform Graph Fourier Transform (GFT) on the remaining attributes, so that signal energies are predominantly projected into the low-frequency coefficients for compact representation. On three standard benchmarks, the method reaches 27x to 33x compression with less than 0.6 dB of PSNR loss, improving on recent post-training compression methods in both size and rendering quality.

Colorectal cancer (CRC) is the second most deadly and third most common cancer, and the leading cause of death among gastrointestinal cancers. Early diagnosis is crucial for the treatment of this cancer and increasing the survival rates. Although CRC is more common in developed regions, its occurrence is also increasing in developing regions as well. CRC diagnosis relies on histopathology assessment post-biopsy. Automated deep learning algorithms can significantly reduce diagnosis time, enhancing efficiency and supporting timely clinical decisions. We present an automated segmentation pipeline for whole-slide histopathology images that labels tumor grades 1-3 and normal mucosa. It utilizes dense prediction transformers with various encoder backbones, overlapping patches, and test-time augmentation. An adaptive augmentation policy, guided by large language models, further improves training. Top models were ensembled via soft voting, and mask refining post-processing steps, Gaussian blurring, morphological closing, and connected components analysis. On a colorectal cancer grade dataset, our method improved the F1 score from 62.92 to 69.84. Code is available here: github.com/caglarmert/ICIP2025

Onboard AI is gaining interest for space applications such as vessel, wildfire, and cloud detection, where real-time processing can improve mission reactivity and reduce downlink needs. However, onboard models may operate on raw or minimally processed imagery rather than on restored ground products. This study evaluates how image degradation affects object detection by varying Signal-to-Noise Ratio (SNR), Modulation Transfer Function (MTF) at Nyquist, and Ground Sampling Distance (GSD). Controlled degradations are applied to Very High Resolution Maxar imagery, and three lightweight detectors, YOLOv5s, YOLOX-S, and NanoDet, are evaluated on the resulting operating points. The results show that the impact of image quality depends on the degradation mechanism, and that increasing degradation does not necessarily lead to a proportional decrease in vessel detection performance. GSD produces the most consistent performance shift, while MTF and SNR effects depend more on the model and resolution. Severe combinations of blur and noise produce the largest losses. These results provide task-level information that can support sensor, processing, and AI trade-offs for future onboard systems.

Estimating biological age is an important task in ageing research, as it quantifies individual ageing trajectories beyond chronological age. Retinal age estimation has become a well-established direction in this area because retinal imaging provides a non-invasive window into neural and microvascular ageing. Existing studies, however, have predominantly focused on fundus photographs and usually model retinal ageing as a single generic process. Although biological ageing is heterogeneous and different organs or tissues may age at different rates, little research has explicitly distinguished neurostructural and vascular ageing in retinal age prediction. This work fills this gap by studying neurostructural ageing with Optical Coherence Tomography (OCT) images and vascular ageing with Optical Coherence Tomography Angiography (OCTA) images. We formulate multimodal OCT/OCTA retinal age estimation as a structural-vascular ageing decomposition problem and propose SAP-DPF, a unified dual-path prediction framework that separately estimates structural and vascular ageing, models modality-specific predictive uncertainty, and adaptively fuses the two ageing signals through an uncertainty-gated late-fusion module. Using this unified prediction framework, SAP-DPF achieves a mean absolute error of 4.07 years for the structural pathway and 7.37 years for the vascular pathway, while the proposed late-fusion algorithm further improves the overall mean absolute error to 4.02 years, representing a 12.52% improvement over the strongest single-modality baseline. This work could extend retinal age prediction beyond a single biological-age estimate, providing a framework for investigating structural and vascular contributions to heterogeneous retinal ageing.

Preoperative liver-tumor assessment requires segmentation, physical-space measurement, visual evidence, and resection planning from the same three-dimensional CT volume. Existing tools often handle these steps separately, while language models cannot reliably compute physical measurements from CT. To provide an integrated workflow, we present VoxelSage, a multi-modal system for two- and three-dimensional visualization, liver-tumor analysis, and preoperative resection planning. Its dual-port architecture separates language-model orchestration from image computation: Port A interprets requests and selects skills, while Port B applies them to CT volumes and segmentation masks and returns structured results. Keeping physical measurements in Port B prevents the LLM from computing them directly and reduces the risk of fabricated numerical results. Eight built-in skills support quantitative analysis, visual evidence generation, three-dimensional reconstruction, segmentation refinement, and sequential resection planning; user-defined skills can extend these functions. For sequence planning, a behavior-cloned neural ranker orders candidate resection targets, while a simulator-based shield checks them against predefined constraints. Across 256 unseen simulator scenes, this approach reduced mean simulated time from 34.274 to 33.388 min (0.886 min, 2.59%) and mean simulated blood loss from 300.847 to 183.852 mL (116.995 mL, 38.89%) relative to a deterministic baseline. These results demonstrate system integration and simulator-level performance, not clinical efficacy or safety. The public implementation is available at https://github.com/ZJUMAI/VoxelSage.

We argue that dividing codec design into Coding for Machines (CfM) and Coding for Humans (CfH) is a misleading distinction for deciding what information a codec may discard. Receiver identity does not determine admissible information loss. The required rate depends on task scope, including the predictions to support, their losses and tolerated risks, the encoder observation, and the permitted decoding procedures. Notably, a machine task may have a higher minimum rate than a restricted human decision. Rate savings on selected machine tasks apply only to the stated requirements, not to an intrinsic ordering by receiver type. We extend source and feature coding to finite task families, derive when restricting the encoder observation preserves the minimum rate, and show that equality between source and split-feature coding rates can no longer hold as the task scope expands.

The use of super-resolution as a post-processing step following a video codec allows videos to be encoded at reduced spatial resolution at the encoder side and to be reconstructed and upsampled to the original resolution at the decoder side. In this way, the required bitrate is reduced and the quality of the reconstructed frames is improved without modifying the core coding architecture. However, generic SR models are typically trained offline and lack sufficient adaptability to the diverse content characteristics and compression artifacts produced by video codecs, which limits their effectiveness during test time. To address the limitation, this paper proposes CASR (Content-Adaptive Super-Resolution), a content-adaptive SR post-filter framework for Versatile Video Coding (VVC), based on encoder-side overfitting on each input sequence. In order to limit the bitrate overhead required for signalling the content adaptation signal, i.e. the weight-update, Low-Rank Adaptation (LoRA) is leveraged. The method freezes the convolution kernels of a pretrained SR network and fine-tunes only lightweight rank-r matrices attached to selected convolution layers, using VVC decoded frames and quantization-parameter (QP) maps of test sequences as supervision. The resulting low-rank update is compressed with the MPEG Neural Network Compression and Representation (NNR) standard. Experiments on the JVET common test conditions (CTC) class A1 and A2 sequences indicate that LoRA-based content adaptation provides bitrate savings over a non-adapted SR post-filter at a small signalling cost. Compared with the VVC Test Model (VTM21), the proposed method achieves BD-rate savings of -10.93% (Y), -15.39% (U), -24.41% (V) under random access and -13.43% (Y), -5.75% (U), -22.94% (V) under all-intra. An ablation of the LoRA rank r further shows that r=4 provides the best trade-off between coding gain and signalling cost.

Few-step streaming audio–video generation requires both causal modeling and step distillation, yet standard training recipes face two context-related challenges. Teacher forcing pairs clean history with a noisy target, but supervises predictive contextual representations only indirectly through velocity prediction. Meanwhile, directly reusing bidirectional score models in causal Distribution Matching Distillation (DMD) creates a mismatch between generation and scoring contexts. We address these challenges with Salt++, a two-stage post-training framework comprising Causal Self-Flow (CSF) and context-aligned autoregressive DMD. CSF exploits contextual information asymmetry by varying the history while keeping the noisy target fixed: a noise-mixed-history student aligns its intermediate representations with those of a clean-history exponential-moving-average teacher. This self-supervised signal encourages the student to extract semantic information and improves cross-modal alignment. Context-aligned AR DMD shares the causal mask and prefix across generator sampling, fake-score training, and real-score evaluation to match generated and reference distributions under a block-conditional KL objective. With calibrated teacher guidance, it performs clean-prefix few-step distillation and then adapts to generated histories without switching objectives or requiring separate consistency distillation. At 480p, Salt++ improves visual and motion quality by 57% and 45% over OmniForcing on JavisBench under the same 4-step causal setting. A separate scale-wise post-training stage extends Salt++ to 4-step $1664\times960$ generation, outperforming bidirectional LTX-2 on six of seven reported metrics. Project page: https://xingtongge.github.io/Saltpp

This paper proposes a ternary visible light communication method using an event-based vision sensor (EVS) and a liquid crystal display (LCD). The throughput of optical camera communication (OCC) is limited by the frame rates of conventional frame-based cameras, and EVSs are expected to overcome this limitation because their pixels asynchronously trigger events with polarity in response to brightness changes, with a temporal resolution on the order of microseconds. However, existing event-based OCC relies on binary signaling that uses only the presence or absence of events and leaves the polarity unused. The proposed method maps the three brightness transitions (increase, decrease, and no change) to ternary symbols, thereby increasing the information carried per symbol. At the transmitter, the LCD displays a marker in which the transmitted data are encoded; at the receiver, the EVS captures the marker and records the brightness changes as events with positive and negative polarities. To suppress errors caused by the timing mismatch between display updates at the transmitter and demodulation at the receiver, we also propose a demodulation algorithm based on frame-transition detection. Experimental results demonstrate that the proposed method achieves a throughput of 336.5 kbps, approximately 150% higher than that of its binary counterpart.

Text-guided image editors can generate high-fidelity medical deepfakes, challenging the reliability of clinical imagery. Although reasoning-based detectors perform strongly in distribution, they degrade substantially under deployment shift. MedForge-Reasoner, an 8B vision-language model trained with supervised fine-tuning and reinforcement learning, achieves 99.2% accuracy on its target distribution, yet misclassifies 40% of authentic scans, reaches only 77% accuracy on unseen generators, and falls to 59% under transmission distortion. Adapting such models through conventional retraining is costly, requiring large-scale supervision and expert-designed guidelines. We introduce MedForge-RSI, a recursive self-improvement framework that enables a deployed detector to adapt while keeping its model weights frozen. Over 20 rounds, the detector analyzes verified errors, accumulates reusable experience, and autonomously develops image-analysis tools, while an independent acceptance test retains only validated improvements. Across 49 registered configurations, MedForge-RSI increases average accuracy over four test sets from 75.0% to 84.4%. On a held-out 4,000-image evaluation, it improves clean accuracy from 76.5% to 87.9% and transmission-distorted accuracy from 59.0% to 70.5%, with the largest gains in authentic-image recall. Controlled analysis across all 49 trajectories identifies which self-improvement mechanisms replicate across seeds, shows acceptance testing to be the largest individual contributor, and reveals a taxonomy of failed adaptations. We release the complete trajectories, including all rejected and rolled-back changes.

Distributed inference has become an indispensable part of deploying medical models under practical latency, memory, and throughput constraints. Although modern frameworks improve serving efficiency through tensor parallelism, mixed precision, kernel fusion, and multi-device communication, they are generally assumed to preserve the behavior observed during centralized HuggingFace evaluation. This assumption creates an evaluation-deployment mismatch: a model may pass offline evaluation but produce a different output after the execution stack changes. To address this mismatch, we propose a testing framework and an improved, distributed-execution-sensitive medical-model benchmark that evaluates the same checkpoint and input under a centralized HuggingFace reference and matched distributed deployments. Extensive experiments across language, vision, and multimodal medical models show that execution changes can produce measurable output disagreements. Across supported visual settings, the test success rate ranges from 0.21 to 0.43 for single-modality models and from 0.32 to 0.98 for multimodal models. The benchmark is aimed at extending medical-model evaluation from capability and security to evaluation-deployment consistency.

Deep learning classifiers for dermoscopic skin lesions often reach high in-distribution accuracy while quietly relying on spurious background cues such as skin tone, device vignetting, and embedded rulers, rather than on lesion morphology. This undermines robustness and fairness across skin tones. This work asks whether Explainable AI (XAI), typically used only to audit a finished model, can instead be repurposed as an active training signal that corrects this shortcut without sacrificing diagnostic accuracy. We introduce CAMEO (Class Activation Mapped Equitable Overlay), a framework that improves skin-lesion classification by selecting stable model explanations and using them to separate lesions from their backgrounds. It then replaces the background with realistic synthetic skin while keeping the lesion unchanged. On HAM10000 and dark-skin ISIC images, CAMEO maintained accuracy while reducing background-driven errors by nearly four times. It also made the model's attention more consistent when backgrounds changed. Results across multiple tests show that reducing reliance on background information improves robustness, with Fitzpatrick-based backgrounds providing a realistic and interpretable approach. Results show that XAI-guided augmentation can make dermoscopic classifiers measurably more robust and fair at no cost to accuracy. They also clarify that it is the mechanism and not the specific tone palette that matters, and that the lasting contribution of XAI here lies in stability-screened, annotation-free lesion localisation rather than in the robustness number itself.

Accurate prediction of post-fire snow conditions does not necessarily establish how wildfire changed those conditions. We developed an explainable geospatial artificial intelligence framework combining multisource Earth observations, meteorological information, and matched before-after comparisons to examine seasonal snow persistence following the 2020 Creek Fire in California's Sierra Nevada. Harmonized Landsat Sentinel-2 observations were used to estimate the fraction of clear-sky observations containing snow during October-July for eleven water years, 2016-2026, on a 500 m analysis grid. We matched 3,778 snow-zone cells inside the fire perimeter to comparable unburned controls using terrain and pre-fire snow conditions; absolute standardized mean differences after matching were no greater than 0.047. HLS persistence agreed closely with MODIS, with a mean annual spatial correlation of 0.947. The landscape-average before-after control-impact contrast was +0.0017, with a year-level 95% confidence interval of -0.023 to +0.027. A stronger response emerged in the highest burn-severity class, where observed persistence increased by 0.026 relative to matched controls, equivalent to 2.6 percentage points. Under 5 km spatial cross-validation, XGBoost predicted raw post-fire persistence with an out-of-fold coefficient of determination of 0.811, whereas predictive skill for the fire-adjusted anomaly reached 0.046. Elevation and temperature together accounted for 65.7% of mean absolute model attribution. These findings distinguish a severity-associated optical snow response from the terrain-climate relationships that dominate predictive skill. Combining matched comparisons with explainable GeoAI provides a practical framework for forest monitoring that separates accurate environmental mapping from inference about disturbance effects.

Power-grid digital twins must combine data-driven prediction with physically meaningful state evolution while preserving the provenance of environmental observations. This paper presents an early-stage EnergyEminence testbed that couples an IEEE 118-bus-style graph-temporal predictor, nonlinear AC cascade simulation, and operator-dashboard-like temporal replay. In addition, we introduce a shared bounded calibration that converts wildfire-detection confidence and spatial extent into source-comparable wildfire interpretable and explainable evidence. We then evaluate it with visually diverse fire and hard-negative videos. Sixteen synthetic environmental videos are curated to generate 160 source-tracked grid scenarios, and a source-video-disjoint test yields 10 true positives, 8 false positives, 22 true negatives, and no false negatives. The errors occur in stressed, non-cascading scenarios conditioned on an unseen growing-fire source. Our diagnostic then reveals environmental shortcut learning that is obscured by scenario-level random splitting. The paper therefore contributes a data-centric and inspectable evaluation workflow for multimodal grid-resilience models, together with evidence supporting separation of environmental alerting from electrical cascade inference

This work presents an optimized compression framework for direct time-of-flight (dToF) light detection and ranging (LiDAR), using an accurate, compact timestamp-encoding strategy to address the high data-rate bottleneck from single-photon timing information in single-photon avalanche diode (SPAD) arrays. We propose a hardware-friendly timestamp-to-depth framework in which a sparse single-photon encoding strategy, namely coarse-to-fine spline sketches (CFSS), projects photon timestamps into fine-grained, low-dimensional representations, called sketch values. CFSS converts timestamps on-the-fly to sketch values and accumulates them with lower memory consumption, without constructing histograms, thereby saving on-chip/device memory. CFSS's closed-form solution for ToF retrieval ensures iteration-free reconstruction. Compared with the conventional sketched-LiDAR framework, the proposed method retains the same compression ratio in the single-peak case, ranging from hundreds$\times$ to thousands$\times$ depending on the accuracy-complexity trade-off, while improving depth-estimation accuracy. Both synthetic and real datasets are used to validate the performance. The CFSS framework is further extended to two-peak scenarios, enabling recovery of objects obscured by partially occluding camouflage and uniformly scattering semi-transparent media. Firmware implementation is also investigated by clearly separating online hardware processing from offline software workloads.

Under stringent bitrate constraints, existing video codecs struggle to balance source fidelity and perceptual realism. Distortion-oriented codecs often oversmooth details, while generative codecs risk introducing content and structural deviations and rely on codec-specific designs. This motivates a question: Can existing codecs achieve a better distortion–perception trade-off through simple, reusable adaptation? Our insight is that native codec reconstructions provide a shared interface through which generative refinement is anchored to source content while remaining decoupled from codec-specific representations. We therefore propose Gen2-VC, a generative video compression framework that enhances the outputs of learned and conventional codecs with a pretrained video prior, leaving their bitstreams and reference update processes unchanged. With the codec, VAE, and generative backbone frozen, lightweight LoRA adapters refine codec reconstruction through single-frame spatial adaptation followed by multi-frame temporal adaptation with the video prior. Using Wan2.1-T2V-1.3B, Gen2-VC-DCVC-UF outperforms previous leading codecs in LPIPS/DISTS and, to our knowledge, is the first generative video codec to surpass VTM-23.0 in both PSNR and MS-SSIM at low bitrates, based on BD-rates averaged over six datasets. Compared to VTM-23.0, it reduces bitrate by an average of 86.65% and 94.24% at matched LPIPS and DISTS, respectively. Adapters trained only on DCVC-UF improve perceptual quality on DCVC-RT, ECM, VTM, and HM without retraining.

Visual perception organizes incomplete sensory evidence into coherent structures, but where do the principles governing this organization come from? We investigate whether learning to predict missing visual information can give rise to structural priors resembling classical Gestalt principles. Masked autoencoders (MAEs) trained on natural images exhibit perceptual completion consistent with good continuation, symmetry, and global form. Critically, when local evidence surrounding a missing region is held constant, manipulating distant elements systematically changes its reconstruction, demonstrating that local predictions are constrained by the global organization of the visual input. We next manipulate visual experience by adapting MAEs to either natural images or texforms that preserve local image statistics while disrupting global spatial organization. This altered experience weakens globally supported completion and shifts perceptual inference toward local continuation when local and global constraints conflict. Together, these results suggest that perceptual organization reflects structural regularities of natural scenes acquired through visual prediction: predictions integrate structural priors across multiple spatial scales, and the balance between local and global organization adapts to the statistical structure of visual experience. More broadly, our findings suggest that classical Gestalt principles may emerge as adaptive perceptual priors learned from the statistical structure of the visual world.

Medical ultrasound (US) is inherently degraded by speckle, a granular interference pattern that is often treated as a complex form of noise in image restoration. However, unlike random noise, US speckle originates from coherent scattering within tissue and is therefore highly spatially dependent and deterministic under fixed acquisition conditions, making US speckle suppression fundamentally different from natural image denoising. Because speckle-free US targets are unavailable in practice, self-supervised denoising is necessary. Blind-spot networks (BSN) are the dominant self-supervised paradigm for natural images, but their pixel-wise masking strategy assumes spatially independent noise, an assumption poorly matched to US speckle, which is spatially correlated over multiple pixels rather than pixel-wise independent. To address this mismatch, we propose Mask2Restore, a self-supervised US despeckling framework that reformulates despeckling as contextual inpainting with block-wise masking on single noisy images. Unlike pixel-wise BSN masking, block-wise masking addresses this multi-pixel speckle correlation by removing locally correlated speckle neighborhoods and shifting the reconstruction cues used by the network from adjacent speckle correlations to broader anatomical context. We further introduce cross-resolution context regularization (CRCR), which suppresses residual speckle bias by enforcing consistency across multi-resolution predictions. Experiments on simulated and in vivo carotid US, unseen fine-structure cases, and downstream cardiac segmentation demonstrate improved speckle-detail trade-offs, better preservation of fine anatomical structures, and practical value for subsequent image analysis.

Gromov-Wasserstein Distances (GWDs) provide quantitative ways of comparing probabilistic distributions defined on different metric spaces by applying techniques from the optimal transport theory. As such, GWD can be potentially useful in a large variety of applications ranging from graph matching problems to 3D object detection. However its practical use at scale is significantly limited by cubic time complexity computations involving dense intra-space distance matrices. Even though in the Euclidean metric spaces several techniques (e.g. involving scalable kernel methods) were proposed to address it, to the best of our knowledge, analogous techniques for general geodesic distances on manifolds, or shortest-path distance on graphs in their discretized variants, were not developed. In this paper, we present Efficient Geodesic Gromov-Wasserstein methods (EGGroW), a new class of efficient algorithms designed to calculate geodesic Gromov-Wasserstein distances with entropic Sinkhorn-like approaches, leveraging recently introduced GenusSink methods \citep{genussink} and the theory of random features. We provide important downstream applications, namely: 3D pose estimation and 3D template detection. In the latter setting, we formulate a partial 3D template recovery as a staged problem: capacity-constrained scene selection is followed by semi-relaxed recovery of template visibility and correspondence. Our empirical findings show that EGGroW provides accurate solutions when standard Euclidean-based techniques fail and is characterized by light computational footprint, as our theoretical analysis predicts.

Video super-resolution (VSR) is an ill-posed inverse problem that aims to reconstruct a high-resolution (HR) video from a noisy, low-resolution (LR) version of it. We present LoCoVSR, a diffusion-based VSR framework that leverages pixel-space denoising diffusion probabilistic models. LoCoVSR integrates the Diffusion Posterior Sampling technique with spatio-temporal context learning, operating in a moving-average form. A localized window of adjacent LR frames is used for recovering each center frame, while applying a shared noise trajectory across all frames. The localized windowing enables processing of long videos without length limitations, supports parallel inference, and prevents error accumulation that may occur in recursive processing. Unlike prior methods, LoCoVSR offers a simple yet very effective VSR solution, avoiding explicit optical flow estimation, or information loss caused by latent space processing. Trained on the VFHQ face dataset, LoCoVSR achieves accurate, temporally consistent and high-quality upscaling with competitive results against recent diffusion-based VSR approaches.

Filmmakers and visual artists routinely need to compose multiple references, actors, locations, props, cultural elements, into a single coherent shot, but existing tools either fail to scale past a handful of references or destroy fine grained subject identity in the process, since per reference tokenization scales memory linearly with reference count $N$ and generated content often departs from the given references rather than reproducing them. We propose RefCompose, a pixel space compositional conditioning framework that decouples where things go from what they look like, via a single fixed resolution reference canvas that keeps conditioning size constant regardless of reference count. Spatial layout is induced at inference time from a frozen diffusion transformer and extracted via Grounding DINO, requiring no LLM or dedicated layout model, while dual stream LoRA adapters inject a layout derived depth map and the encoded canvas through separate low rank streams, disentangling geometric scaffolding from localized appearance. On the Dense Layout protocol, RefCompose consistently outperforms layout based and state of the art multi reference baselines on color, texture, shape, spatial accuracy, and identity/content preservation at higher reference counts, all with constant inference memory, making it a practical building block for multi subject cinematic composition at production scale.

Acquiring data is costly: higher measurement fidelity costs power and storage and risks collecting irrelevant content, while aggressive cost reduction can discard information that later analysis needs. We address this trade-off with an information-theoretic framework that acquires data relevant to a broad set of tasks rather than to one model. A mask policy, conditioned on side information, chooses which pixels to measure so as to maximize the mutual information between a discrete image and its partial observation under a budget; since the image entropy does not depend on the mask, this is equivalent to minimizing the conditional entropy. A frozen discrete denoising diffusion model (D3PM) supplies the posterior, and we use it in two ways: as an entropy surrogate for training a one-shot mask generator, and as the criterion for sequential greedy acquisition. The one-shot generator outperforms random masks only with care, including an unbiased gradient estimator for binary masks. With sequential acquisition, on MNIST the prior makes $8\times$ fewer errors than random at a $10\%$ budget, and on CIFAR-10 it gains $0.9$–$3.4$~dB. On fastMRI, our proposed technique using a static mask outperforms the well-known methods such as variable density and LOUPE.

High-resolution plant imaging enables detailed characterization of plant morphology, but dense scientific analysis remains limited by costly pixel-level annotations, large image pixel dimensions, and substantial variation in imaging conditions. This work proposes a scalable in-domain self-supervised pretrained foundation model for high-resolution, high-pixel-dimension multi-species plant imagery. A masked autoencoder with a ViT backbone is pretrained on more than 10 million multi-view plant image tiles using distributed training. Following scalable pretraining, the learned foundation-model representations are comprehensively benchmarked across fine-grained dense prediction and coarse global feature recognition, with particular emphasis on limited supervision and realistic downstream imaging conditions. This work focuses on the domain gap of existing foundation models in dense feature representation and transfer. Through extensive experiments involving limited annotations, cross-view variation, and resolution degradation, the in-domain FM achieves a Mean Dice of 0.8686 and a Pooled Dice of 0.8959, outperforming an MAE counterpart pretrained on large-scale natural-image data by 0.0694 and 0.0613, respectively. The results further indicate that increasing pretraining scale produces consistent improvements in dense feature transfer. Overall, these findings suggest that scaling in-domain self-supervised pretraining can reduce the domain gap and improve transferable dense representations for high-pixel-dimension scientific imaging.