SEARCH · Searcharxiv

Search Searcharxiv

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources

arXiv2026

CylindTrack: Depth-Aware Cylindrical Motion Modeling for Panoramic Multi-Object Tracking

Multi-Object Tracking (MOT) is essential for embodied perception, where panoramic cameras provide a 360° field of view to reduce blind spots and keep surrounding targets observable for longer durations. However, panoramic MOT requires motion estimates and association cues consistent with its imaging geometry. In videos represented in equirectangular projection (ERP), horizontal coordinates wrap at the image seam, while frame-wise relative-depth estimates may fluctuate over time. These inconsistencies can disrupt trajectory prediction and identity association. To address these challenges, we propose CylindTrack, a depth-aware cylindrical tracking-by-detection framework that integrates continuous angular motion modeling with temporally filtered relative-depth cues. Specifically, Depth-Temporal Trajectory Modeling (DTM) promotes instance depth from an isolated frame-wise cue into temporally filtered trajectory-level states. Spherical Spatio-Temporal Consistency Learning (SSTC) combines a Temporal Mixer and Spherical Geometry-aware Attention to enhance temporal coherence and panoramic geometric alignment in depth-aware representations. Furthermore, a Topology-Aware Cylindrical Motion Model (TCMM) lifts horizontal motion into a continuous angular state space, enabling seam-consistent motion prediction and association. By jointly modeling trajectory-level depth consistency and panoramic topology, CylindTrack improves identity preservation and trajectory continuity in challenging panoramic scenes. Experiments on QuadTrack and JRDB demonstrate improvements in identity-oriented metrics with practical efficiency, achieving HOTA scores of 33.67 and 31.12, IDF1 scores of 40.45 and 34.33, and tracking speeds of 28.56 and 21.34 FPS, respectively. The source code will be released at https://github.com/warriordby/CylindTrack.

cs.CV↗

arXiv2026

Fast Trainable Multilinear Bases for Image Compression

The Discrete Fourier Transform (DFT), the Discrete Cosine Transform (DCT), and their block-wise variants underpin most deployed image and video codecs. Their effectiveness rests on three properties: their runtime is near-linear (up to a polylogarithmic factor) in the image size, they are exactly invertible, and they carry few to no parameters. We generalize these transforms to dataset-adaptive isometric multilinear bases that keep all three properties at the cost of a polylogarithmic number of parameters, by parameterizing each basis as an isometric tensor network, inspired by quantum many-body theory, and training one shared basis per image dataset with Riemannian optimization. Because the family contains the DFT and the DCT-IV (a variant of the DCT) as exact points, training can start from these transforms, and it improves on its starting point at the compression level it is trained for. We evaluate the bases on natural photographs and line drawings. On photographs, which the cosine basis already fits well, the best trained basis leads the DCT only under aggressive truncation. On Quick Draw line drawings, under a common codec at matched quality, the best trained basis stores each image in about 20% fewer bytes than the 8x8 block cosine transform of the Joint Photographic Experts Group (JPEG) format, not counting the one-off storage of the basis.

eess.IV↗

arXiv2026

Adaptive Fused Prior Transfer for Controllable Generative Image Compression

At very low bitrates, image compression discards fine textures and local structures, while distortion-oriented reconstruction often produces over-smoothed images. Generative codecs synthesize missing details, but existing codebook-based controllable designs generally rely on single-codebook reconstruction priors. We propose Adaptive Fused Prior Transfer for Controllable Generative Image Compression (AFP-GIC), which transfers an image-adaptive fused prior from a frozen pretrained AdaCode model. Encoder-side prior features guide latent formation, while the decoder predicts a compatible fused prior from the compressed representation and control variables, without transmitting the prior itself. A motivating analysis shows that better decoder-side prior alignment tightens a reconstruction-error upper bound and that the fused-prior family includes single-codebook choices as special cases. A single pretrained model supports five evaluated bitrate operating points. Under the unified benchmark, AFP-GIC achieves 18.1% lower decoder latency and uses 31.10 million (20.5%) fewer inference parameters than DC-VIC. Experiments on Kodak, CLIC2020, and DIV2K show competitive PSNR and SSIM, with the clearest naturalness gains in NIQE scores and very-low-bitrate visual comparisons. Code: https://github.com/yifeipet/AFP_GIC.

eess.IV↗

arXiv2026

Feature-Spectral Fragility in Segmentation: Dataset Dependence, Architecture-Specific Localization, and Spectral Correlates

Robustness of segmentation models is commonly assessed through input-domain perturbations, while dependence on frequency content within learned feature representations remains less understood. We probe this dependence using targeted post-training low-pass interventions on internal representations of three segmentation architectures, ResNet50-UNet (CNN), VM-UNet (SSM), and Swin-UNETR (Transformer), across CVC-ClinicDB and ISIC2018, with headline evaluations performed on untouched held-out test sets. At cutoff $ρ=0.25$, feature-domain low-pass filtering causes severe degradation on CVC: Dice drops by 100%, 73.2%, and 30.9% for CNN, SSM, and Transformer, respectively, compared with 9.4%, 10.3%, and 0.6% on ISIC. The cross-dataset difference is statistically significant for every architecture. Single-stage interventions further show that the most sensitive stage depends on architecture and dataset: the CNN peaks at the 2nd to 3rd encoder block, whereas the SSM peaks at the 1st to 2nd encoder stage. Native feature-domain spectral measurements show an inverse association between high-frequency energy and fragility on CVC; the relationship is only partial on ISIC and is therefore treated as a candidate correlate rather than a proven mechanism. Finally, Fourier augmentation improves robustness to input-space low-pass filtering but leaves feature-domain degradation essentially unchanged. These results show that feature-spectral robustness is strongly dataset-dependent, architecture-specific, and distinct from input-domain spectral robustness.

cs.CV↗

arXiv2026

VideoPulse: Neonatal heart rate and peripheral capillary oxygen saturation (SpO2) estimation from contact free video

Remote photoplethysmography (rPPG) enables contact free monitoring of vital signs and is especially valuable for neonates, since conventional methods often require sustained skin contact with adhesive probes that can irritate fragile skin and increase infection control burden. We present VideoPulse, a neonatal dataset and an end to end pipeline that estimates neonatal heart rate and peripheral capillary oxygen saturation (SpO2) from facial video. VideoPulse contains 157 recordings totaling 2.6 hours from 52 neonates with diverse face orientations. Our pipeline performs face alignment and artifact aware supervision using denoised pulse oximeter signals, then applies 3D CNN backbones for heart rate and SpO2 regression with label distribution smoothing and weighted regression for SpO2. Predictions are produced in 2 second windows. On the NBHR neonatal dataset, we obtain heart rate MAE 2.97 bpm using 2 second windows (2.80 bpm at 6 second windows) and SpO2 MAE 1.69 percent. Under cross dataset evaluation, the NBHR trained heart rate model attains 5.34 bpm MAE on VideoPulse, and fine tuning an NBHR pretrained SpO2 model on VideoPulse yields MAE 1.68 percent. These results indicate that short unaligned neonatal video segments can support accurate heart rate and SpO2 estimation, enabling low cost non invasive monitoring in neonatal intensive care.

eess.IV↗

arXiv2026

Medical Image Segmentation based on Deep Active Contour and Mean Curvature Loss Function

Medical image segmentation is an important task in clinical analysis. Although deep learning techniques are widely used, training at the individual pixel level ignores geometric prior information about the region being segmented. Integrating the Chan-Vese model into the loss function is a well-established remedy that accounts for the regions inside and outside the segmentation and, through its length term, for boundary regularity. However, such losses still lack an effective characterisation of local boundary geometry. We introduce the mean curvature as a natural geometric constraint and propose a Deep Active Contour and Mean Curvature (DACMC) loss function, in which a fixed convolution kernel approximates the mean curvature at negligible computational cost. The loss has a single hyper-parameter, the curvature weight $λ$, fixed at $10^{-3}$ for all experiments. We evaluate DACMC on three public datasets - liver computed tomography (CT), spleen magnetic resonance imaging (MRI) and dermoscopy images from the International Skin Imaging Collaboration (ISIC) - using two encoder-decoder networks as backbones and the Dice similarity coefficient (DSC), the 95th-percentile Hausdorff distance (HD95), the Jaccard similarity (JS) and the average surface distance (ASD) as metrics, against the cross-entropy, Dice, active contour and elastica losses. DACMC attains the best or second-best DSC in five of the six dataset-backbone settings; on spleen MRI it reduces HD95 to 16.28 millimetres and ASD to 1.90 millimetres, and on ISIC it reduces HD95 to 7.08 millimetres. A sensitivity study shows a broad plateau for $λ\leq 10^{-3}$ and degeneration only when the curvature term dominates.

eess.IV↗

arXiv2026

Spheriverse: 3D Scene Understanding from Spherical Observations in the Wild

Spherical observations provide global visual context for 3D scene understanding. However, visual information is encoded in an angular domain, whereas the physical world is represented in Cartesian coordinates. This cross-space representation gap complicates geometric correspondence and semantic evidence aggregation. To delve into this challenge, we introduce Spheriverse, comprising 64,400 temporally aligned spherical image-LiDAR pairs organized into 644 sequences. The dataset spans diverse scenes, illumination, and weather conditions, with fine-grained semantic classes. We further establish benchmarks for semantic occupancy prediction, semantic mapping, and 3D object detection, evaluating 30+ methods through overall and scene-wise comparisons. For dense prediction, we propose SphereOcc, an occupancy framework that couples spherical geometry modeling with semantic evidence retrieval. Cartesian-Spherical Representation Remodeling (CSRR) incorporates spherical range-azimuth geometry into Cartesian voxel features through region-wise modulation. Spherical Evidence Re-querying (SER) then conditions queries on voxel content and range-height-azimuth geometry to adaptively retrieve relevant semantic evidence from source spherical image features. SphereOcc achieves 13.91% mIoU and 24.65% GeoIoU, yielding relative improvements of 13.9% and 9.3% over the respective best-performing methods, TPVFormer and SurroundOcc. It also ranks first in both metrics across all five scene categories, with consistent advantages across the evaluated spatial partitions and reduced fields of view. The established benchmark and source code will be available at https://feit-feiteng.github.io/Spheriverse.

cs.CV↗

arXiv2026

L2G-Map: Local-to-Global Mapping via Hierarchical Diffusion Refinement and Elliptical Bayesian Fusion

Offline high-definition maps provide essential geometric and topological priors for autonomous driving systems. Pure-vision solutions have become the predominant paradigm for offline mapping due to their cost-effectiveness and scalability. However, local-to-global mapping under visual conditions confronts two fundamental challenges: single-shot local observations are susceptible to viewpoint variation and environmental interference, leading to geometric deviations, while multi-source local information exhibits heterogeneous confidence, rendering globally consistent aggregation difficult. To address these, this paper proposes L2G-Map, a framework comprising hierarchical prior diffusion refinement and elliptical space Bayesian fusion. The former jointly embeds temporal context and centerline priors to guide structure completion and topology recovery during denoising, alleviating the information incompleteness inherent in pure-vision settings. The latter incorporates an adaptive weighting strategy driven by elliptical distance propagation, enabling probabilistically optimal aggregation of multi-source information under the Bayesian posterior update paradigm. Extensive experiments on nuScenes and Argoverse benchmark datasets verify the effectiveness of L2G-Map. The proposed refinement component yields consistent local map accuracy improvements across different datasets. Under sensor-degraded conditions, a 3.27% mIoU gain is achieved. Furthermore, the adaptive fusion component significantly enhances the accuracy of global maps. The fused global map can be flexibly embedded into different online map models, yielding an 18.26% mIoU improvement in semantic map construction and a 20.00% enhancement in vectorized map construction, demonstrating the overall advantages of the proposed closed-loop pipeline. Source code will be available at https://github.com/lynn-yu/L2G-Map.

cs.CV↗

arXiv2026

Leveraging Cardiac Imaging to Improve ECG-Based Detection of Chagas Disease in Resource-Constrained Settings

Chagas disease is a major cause of cardiomyopathy in Latin America. Cardiac magnetic resonance (CMR) imaging can characterize its structural abnormalities, but scanners and expert readers remain scarce in endemic regions. Electrocardiography (ECG) is inexpensive and widely available, yet structural disease must be inferred indirectly from electrical signals. We propose to transfer CMR-derived structural knowledge to ECG through contrastive pre-training. Using 63,193 paired ECG-CMR examinations from the UK Biobank, we align an ECG encoder with a clinically grounded CMR embedding space using an asymmetric InfoNCE objective. Despite seeing no Chagas cases during pre-training, the resulting representation improves ECG-based Chagas detection. Across CODE-15% and SaMi-Trop, a frozen linear probe achieves an AUROC of 0.851 and sensitivity at the top 5% of predicted risk (Top5%-TPR) of 0.427 in five-fold cross-validation, compared with 0.827 and 0.377 for an unaligned ECG-FM baseline. On the PhysioNet/CinC 2025 Challenge test set, our model obtains the highest AUROC on SaMi-Trop-3 and the best ELSA-Brasil challenge score among the three top-performing methods, indicating that imaging-supervised ECG representations can generalize to populations and settings beyond the pre-training distribution.

cs.LG↗

arXiv2026

Prototyping QoE-Aware Rate Adaptation in Cellular Networks with Commercial Applications

Prior work has shown that QoE-aware resource sharing for real-time interactive video can support up to three times more simultaneous sessions at acceptable quality compared to rate-fair allocation. However, the required capabilities (QoE-targeted encoding, runtime spatial complexity estimation, and rich application-network APIs) are not yet available in commercial deployments. In this paper, we take an evolutionary approach: we design a system that delivers QoE-aware resource allocation using only capabilities that can be assembled in a lab today. We extend the utility-based allocation framework to the radio resource domain by introducing composite spatial complexity, which combines a session's video spatial complexity with its time-variant spectral efficiency into a single resource demand function. To operate with commercial real-time video streaming applications that use rate-based congestion control and lack capability to measure QoE, we use external tooling for QoE measurements. We develop an incremental reallocation algorithm with per-interval limits that encode both the congestion control algorithm's speed constraint and that spatial complexity estimates are reliable only near the current rate. The resulting prototype combines external QoE measurements with congestion-signal-based rate steering and does not require modification to commercial applications. We chart an evolution path from this prototype toward full QoE-aware resource sharing, mapping emerging standards (IETF SCONE, CAMARA, Media over QUIC) to the progressive capabilities they enable.

cs.NI↗

arXiv2026

NeuralLVC: Neural Lossless Video Compression via Masked Diffusion with Temporal Conditioning

While neural lossless image compression has advanced significantly with learned entropy models, lossless video compression remains largely unexplored in the neural setting. We present NeuralLVC, a neural lossless video codec that combines masked diffusion with an I/P-frame architecture for exploiting temporal redundancy. Our I-frame model compresses individual frames using bijective linear tokenization that guarantees exact pixel reconstruction. The P-frame model compresses temporal differences between consecutive frames, conditioned on the previous decoded frame via a lightweight reference embedding that adds only 1.3% trainable parameters. Group-wise decoding enables controllable speed-compression trade-offs. Our codec is lossless in the input domain: for video, it reconstructs YUV420 planes exactly; for image evaluation, RGB channels are reconstructed exactly. Experiments on 9 Xiph CIF sequences and on 24 HD sequences at 720p and 1080p (Xiph, UVG, and the modern BVI-AOM benchmark) show that NeuralLVC outperforms H.264 and H.265 lossless by a significant margin at every resolution tested. We verify exact reconstruction through end-to-end encode-decode testing with arithmetic coding. These results suggest that masked diffusion with temporal conditioning is a promising direction for neural lossless video compression.

eess.IV↗

arXiv2026

AstraMoE-SR: Trajectory-Guided Diffusion for Blind Satellite Jitter Deblurring and Super-Resolution

Pushbroom satellite imaging couples limited spatial resolution with platform attitude instability. Platform jitter produces spatially varying motion blur because each scan line is acquired under a different instantaneous attitude, while perspective geometry causes the same perturbation to induce different pixel displacements across the field of view. Existing blind restoration methods that assume a spatially invariant kernel and satellite jitter correction methods that rely on auxiliary observations are therefore not directly applicable. We present AstraMoE-SR, a single-image framework that jointly restores motion blur and spatial resolution without auxiliary measurements. Rather than estimating a blur kernel, we infer how the camera moved by reparameterizing degradation as a local exposure trajectory under pushbroom geometry. A conditional diffusion model estimates the trajectory distribution, mitigating the over-smoothing of high-frequency jitter by deterministic point estimation. The predicted trajectory conditions a pretrained latent diffusion backbone through trajectory-guided geometric alignment and spatially adaptive reconstruction. We further show that the remaining point-wise trajectory error is consistent with intrinsic jitter-phase ambiguity that is not resolved by increasing estimator capacity. On all 1,411 DOTA-v1.0 images degraded using our physically motivated forward model, AstraMoE-SR is the only evaluated method to outperform the no-restoration baseline across every fidelity metric, improving on StableSR by 0.64 dB PSNR, 15.2% LPIPS, and 0.091 DINO feature similarity. Reconstructions conditioned on predicted trajectories differ negligibly from those using ground-truth trajectories, indicating that the estimates retain the degradation information required for effective restoration.

eess.IV↗

arXiv2026

Spatial Feature-wise Linear Modulation (SpFiLM) for Contrast Agent-Aware Brain Parcellation

Most automated brain parcellation tools are developed and validated on T1-weighted (T1w) MRI. Yet, some clinical workflows for which parcellation is relevant only use contrast-enhanced T1w (T1ce) MRI, on which T1w-trained models are less accurate. We present a unified network that parcellates both pre- and post-contrast agent T1w MRI reliably, trained on a combination of the two with conditioning that spatially modulates its response differently for each. Feature-wise Linear Modulation (FiLM) is a known approach for input-based modulation in networks. It applies a per-channel scale and shift uniformly across the input. However, the appearance change between pre- and post-contrast varies locally across the brain, making FiLM suboptimal for our use case. In this work, we introduce Spatial FiLM (SpFiLM), a conditioning layer whose modulation varies spatially, assembling a voxel-wise scale and shift from image-derived spatial patterns. Using a cohort of 134 patients with paired T1w and T1ce MRI parcellated into 106 classes, the addition of SpFiLM layers in a UNet increased the mean Dice on the test set of 25 patients from 80.2% to 84.1%, a 4.9% relative improvement. Adding SpFiLM layers led to the best performance on both pre- and post-contrast MRI, even when controlling for network parameter counts.

eess.IV↗

arXiv2026

Cross-modal learning for SAR target recognition using optical vision foundation models

Synthetic Aperture Radar (SAR) is an important modality in a wide range of imaging applications due to its versatile, long range and near all weather operating capabilities. However, Automatic Target Recognition (ATR) remains a challenging problem due to limited labelled data, the strong speckle in SAR images and the significant domain gap between SAR and more abundant optical imagery. In contrast, electro-optical (EO) imagery benefits from massive datasets, clearer visual structure and powerful foundation models. In this work, we investigate how vision foundation models trained on optical data can provide class level supervision for SAR classification. We propose a cross-modal EO to SAR prototype alignment framework in which a frozen EO encoder, based on a DINOv3 vision foundation model, is used to construct class level optical prototypes without requiring strict EO/SAR pairs. A SAR model is then trained to classify SAR images while aligning its embeddings to the corresponding EO class prototype. At inference time, the SAR model operates independently, without access to optical imagery. We evaluate our approach on the UNICORNv2 dataset, an EO and SAR dataset of civilian vehicles with heavily speckled images and severe class imbalance. EO prototype alignment improves SAR classification accuracy over frozen DINOv3, SAR only finetuning and unpaired distribution alignment baselines, and t-SNE visualizations provide qualitative evidence of clearer separation among classes in the trained SAR embedding space. These results suggest that optical vision foundation models, despite being trained on visible spectrum imagery, provide transferable information for SAR image classification, offering a practical method for using large scale pretrained vision foundation models across challenging sensing modalities.

cs.CV↗

arXiv2026

Foundation Models for Generalizable Semantic and Goal-Oriented Communication

Semantic and goal-oriented communication is increasingly studied for 6G, but generalization beyond seen data remains a key weakness under tight rate budgets. Many existing systems overfit their training data and degrade sharply at very low bit rates because they attempt to compress the entire signal. We introduce Foundation Model-Guided Semantic and Goal-Oriented Communication (FMSGOC), a framework that uses broad visual-linguistic Foundation Model priors to mitigate overfitting. It further improves rate efficiency by concentrating bits on sparse, goal-aligned anchors and relying on generative foundation-model priors to reconstruct the masked regions. By decoupling what to send from how to reconstruct, a vision-language foundation model selects and transmits a sparse set of semantic anchors, while a pretrained diffusion model, fine-tuned for masked completion, reconstructs the image at the receiver. In our experiments, FMSGOC reaches 0.039 bits per pixel (BPP), maintains high semantic fidelity (cosine similarity 0.87-0.90 on CIFAR-10), remains robust on previously unseen inputs (0.83-0.86 on ImageNet), and shows good perceptual similarity (0.1278/0.1558, CIFAR-10/ImageNet), outperforming strong end-to-end baselines at lower bit rates.

cs.LG↗

arXiv2026

Toward Robust Single-Photon Perception for Robots: A Condition-Aware Active Learning Approach

LiDAR-based perception plays a fundamental role in modern robotic systems for environment understanding and navigation. Single-photon LiDAR (SPL) extends conventional LiDAR by enabling photon-efficient 3D sensing under challenging conditions such as long-range operation, low-albedo targets, and limited signal returns. However, developing SPL perception models for real-world robotic applications remains difficult because annotated SPL data are costly to obtain and model performance can vary substantially across imaging conditions. In this paper, we present the first active learning framework tailored to the SPL sensing modality rather than a specific downstream task. Our method introduces a physics-grounded, imaging condition-aware sampling strategy that uses synthetic SPL variants to characterize how candidate samples respond to changes in sensing conditions. By jointly modeling prediction uncertainty, sample diversity, and sensitivity to photon-level imaging variations, the proposed approach prioritizes samples that are informative for improving labeling efficiency and robustness. Extensive experiments on synthetic and real-world SPL datasets demonstrate that our method substantially reduces annotation requirements while maintaining strong performance across image-level and dense prediction settings. On synthetic data, our approach achieves 97% classification accuracy using only 1.5% labeled samples. On real-world data, it maintains 90% accuracy with 8.2% labeled samples, outperforming the strongest baseline by 6%. Segmentation results further show that the same condition-aware acquisition principle improves annotation efficiency and robustness across imaging conditions. These results establish a modality-aware active learning strategy for data-efficient SPL perception, with the potential to extend to a broader range of downstream tasks.

eess.IV↗

arXiv2026

SAGE: Shape-Adapting Gated Experts for Adaptive Histopathology Image Segmentation

The significant variability in cell size and shape continues to pose a major obstacle in computer-assisted cancer detection on gigapixel Whole Slide Images (WSIs), due to cellular heterogeneity. Current CNN-Transformer hybrids use static computation graphs with fixed routing. This leads to extra computation and makes it harder to adapt to changes in input. We propose Shape-Adapting Gated Experts (SAGE), an input-adaptive framework that enables dynamic expert routing in heterogeneous visual networks. SAGE reconfigures static backbones into dynamically routed expert architectures via a dual-path design with hierarchical gating and a Shape-Adapting Hub (SA-Hub) that harmonizes feature representations across convolutional and transformer modules. Embodied as SAGE with ConvNeXt and Vision Transformer UNet (SAGE-ConvNeXt+ViT-UNet), our model achieves a Dice score of 95.23% on EBHI, DSC scores of 92.78% and 91.42% on GlaS Test A and Test B, respectively, and 91.26% DSC at the WSI level on DigestPath, while exhibiting robust generalization under distribution shifts by adaptively balancing local refinement and global context. SAGE establishes a scalable foundation for dynamic expert routing in visual networks, thereby facilitating flexible visual reasoning. Project page: https://oxyzgiahuy.github.io/sage/

eess.IV↗

arXiv2026

MedGSSR: Generalizable Medical Image Super-Resolution 3D Reconstruction via Hierarchical Feed-forward Gaussian Splatting

High-resolution volumetric medical imaging is critical for clinical diagnosis, yet acquisition is often limited by scanner hardware, scan time, and for CT, radiation dose. Medical 3D Super-Resolution (Med3DSR) offers a computational alternative, but existing methods commonly rely on per-subject optimization, pretrained priors, or coordinate-based implicit representations, which compromise anatomical fidelity and limit efficiency. To address these limitations, we present MedGSSR, a fully end-to-end feed-forward framework that represents volumes as an explicit 3D Gaussian field for Med3DSR. Unlike coordinate-based implicit functions, our explicit 3D Gaussian representation naturally enhances signal continuity and local high-frequency fidelity. Specifically, MedGSSR explicitly decouples the reconstruction process into coarse-grained structural preservation and fine-grained textural refinement through the proposed Pyramid Anatomical Encoder and a Hierarchical Gaussian Projector. To support arbitrary-scale super-resolution, we introduce sub-voxel Gaussian decomposition and a Differentiable Gaussian Voxelizer that directly queries the continuous 3D intensity field, reducing discretization artifacts. Extensive experiments on MRI and CT benchmarks demonstrate that MedGSSR significantly outperforms state-of-the-art methods. Notably, our framework exhibits robust generalizability across unseen datasets without requiring per-subject optimization, enabling fast inference and high-fidelity volumetric super-resolution in practical clinical settings. Our project webpage, including code, is at https://william2ai.github.io/medgssr

eess.IV↗