Audio and Speech Processing
Speech Recognition Model Improves Text-to-Speech Synthesis using Fine-Grained Reward
Guansu Wang and Peijie Sun
Recent advances in text-to-
speech (TTS) have enabled models to clone arbitrary unseen speakers and synthesize high-quality, natural-sounding
speech. However, evaluation methods lag behind: typical mean opinion score (MOS)
estimators perform regression over entire utterances, while failures usually occur in a few problematic words. We observe that encoder-decoder ASR models (e.g., Whisper) surface word-level mismatches between
speech and text via cross-
attention, providing a fine-grained reward signal. Building on this, we introduce Word-level TTS
Alignment by ASR-driven Attentive Reward (W3AR). Without explicit reward annotations, W3AR uses
attention from a pre-trained ASR model to drive finer-grained
alignment and optimization of sequences predicted by a TTS model. Experiments show that W3AR improves the quality of existing TTS systems and strengthens zero-shot robustness on unseen speakers. More broadly, our results suggest a simple recipe for generative modeling: understanding models can act as evaluators, delivering informative, fine-grained feedback for optimization.
Read the full abstract ▾
Signal Processing
Semantic-driven Wireless Environment Knowledge Representation for Efficiency-Accuracy Balanced Beam Prediction in Vehicular Networks
Jialin Wang, Jianhua Zhang, Yu Li et al. (+2 more)
The rapid
evolution of the internet of vehicles demands ultra-reliable low-latency communication in high-mobility environments, where conventional beam prediction methods suffer from high-dimensional inputs, prolonged training times, and limited interpretability. To address these challenges, the propagation environment semantics-aware wireless environment knowledge beam prediction (PES-WEKBP) framework is proposed. PES-WEKBP pioneers a novel electromagnetic (EM)-grounded knowledge
distillation method, transforming raw visual data into an ultra-lean, interpretable material and location-related wireless environment knowledge
matrix. This
matrix explicitly encodes critical propagation environment semantics, which is material EM properties and spatial relationships through a physics-informed parameterization process, distilling the environment and channel interplay into a minimal yet information-dense representation. A lightweight decision network then leverages this highly compressed knowledge for low-complexity beam prediction. To holistically evaluate the performance of PES-WEKBP, we first design the prediction consistency-efficiency index (PCEI), which combines prediction accuracy with a stability-penalized logarithmic training time to ensure a balanced optimization of reliability and computational efficiency. Experiments validate that PES-WEKBP achieves a 99.75% to 99.96% dimension reduction and improves accuracy by 5.52% to 8.19%, which outperforms state-of-the-art methods in PCEI scores across diverse vehicular scenarios.
Read the full abstract ▾
All authors Jialin Wang, Jianhua Zhang, Yu Li, Yutong Sun, Yuxiang Zhang
Image and Video Processing · Cross-listed from cs.CV
Zhou Xu, Qi Wang, Yuxiao Yang et al. (+3 more)
Score
Distillation Sampling (SDS) enables 3D asset generation by distilling priors from pretrained 2D text-to-image
diffusion models, but vanilla SDS suffers from over-saturation and over-smoothing. To mitigate this issue, recent variants have incorporated negative
prompts. However, these methods face a critical trade-off: limited texture optimization, or significant texture gains with shape distortion. In this work, we first conduct a systematic analysis and reveal that this trade-off is fundamentally governed by the utilization of the negative
prompts, where Target Negative
Prompts (TNP) that embed target information in the negative
prompts dramatically enhancing texture realism and fidelity but inducing shape distortions. Informed by this key insight, we introduce the Target-Balanced Score
Distillation (TBSD). It formulates generation as a multi-objective optimization problem and introduces an adaptive strategy that effectively resolves the aforementioned trade-off. Extensive experiments demonstrate that TBSD significantly outperforms existing state-of-the-art methods, yielding 3D assets with high-fidelity textures and geometrically accurate shape.
Read the full abstract ▾
All authors Zhou Xu, Qi Wang, Yuxiao Yang, Luyuan Zhang, Zhang Liang, Yang Li
Image and Video Processing · Cross-listed from cs.CV
Task-Aware 3D Affordance Segmentation via 2D Guidance and Geometric Refinement
Lian He, Meng Liu, Qilang Ye et al. (+3 more)
Understanding 3D scene-level affordances from natural
language instructions is essential for enabling
embodied agents to interact meaningfully in complex environments. However, this task remains challenging due to the need for semantic
reasoning and spatial grounding. Existing methods mainly focus on object-level affordances or merely lift 2D predictions to 3D, neglecting rich
geometric structure information in point clouds and incurring high computational costs. To address these limitations, we introduce Task-Aware 3D Scene-level Affordance
segmentation (TASA), a novel geometry-optimized framework that jointly leverages 2D semantic cues and 3D
geometric reasoning in a coarse-to-fine manner. To improve the affordance detection efficiency, TASA features a task-aware 2D affordance detection module to identify manipulable points from
language and visual inputs, guiding the selection of task-relevant views. To fully exploit 3D
geometric information, a 3D affordance refinement module is proposed to integrate 2D semantic priors with local 3D geometry, resulting in accurate and spatially coherent 3D affordance masks. Experiments on SceneFun3D demonstrate that TASA significantly outperforms the baselines in both accuracy and efficiency in scene-level affordance
segmentation.
Read the full abstract ▾
All authors Lian He, Meng Liu, Qilang Ye, Yu Zhou, Xiang Deng, Gangyi Ding
Image and Video Processing · Cross-listed from cs.CV
EPSegFZ: Efficient Point Cloud Semantic Segmentation for Few- and Zero-Shot Scenarios with Language Guidance
Jiahui Wang, Haiyue Zhu, Haoren Guo et al. (+3 more)
Recent approaches for few-shot 3D point cloud semantic
segmentation typically require a two-stage learning process, i.e., a pre-training stage followed by a few-shot training stage. While effective, these methods face overreliance on pre-training, which hinders model flexibility and adaptability. Some models tried to avoid pre-training yet failed to capture ample information. In addition, current approaches focus on visual information in the support set and neglect or do not fully exploit other useful data, such as textual annotations. This inadequate utilization of support information impairs the performance of the model and restricts its zero-shot ability. To address these limitations, we present a novel pre-training-free network, named Efficient Point Cloud Semantic
Segmentation for Few- and Zero-shot scenarios. Our EPSegFZ incorporates three key components. A Prototype-Enhanced Registers
Attention (ProERA) module and a Dual Relative Positional Encoding (DRPE)-based cross-
attention mechanism for improved feature extraction and accurate query-prototype correspondence construction without pre-training. A
Language-Guided Prototype
Embedding (LGPE) module that effectively leverages textual information from the support set to improve few-shot performance and enable zero-shot inference. Extensive experiments show that our method outperforms the state-of-the-art method by 5.68% and 3.82% on the S3DIS and ScanNet
benchmarks, respectively.
Read the full abstract ▾
All authors Jiahui Wang, Haiyue Zhu, Haoren Guo, Abdullah Al Mamun, Cheng Xiang, Tong Heng Lee
Image and Video Processing
DualVision ArthroNav: Investigating Opportunities to Enhance Localization and Reconstruction in Image-based Arthroscopy Navigation via External Cameras
Hongchao Shu, Lalithkumar Seenivasan, Mingxu Liu et al. (+6 more)
Arthroscopic procedures can greatly benefit from
navigation systems that enhance spatial awareness, depth perception, and field of view. However, existing optical tracking solutions impose strict workspace constraints and disrupt
surgical workflow.
Vision-based alternatives, though less invasive, often rely solely on the
monocular arthroscope camera, making them prone to drift, scale ambiguity, and sensitivity to rapid motion or occlusion. We propose DualVision ArthroNav, a multi-camera arthroscopy
navigation system that integrates an external camera rigidly mounted on the arthroscope. The external camera provides stable visual odometry and absolute localization, while the
monocular arthroscope
video enables dense scene reconstruction. By combining these complementary views, our system resolves the scale ambiguity and long-term drift inherent in
monocular SLAM and ensures robust relocalization. Experiments demonstrate that our system effectively compensates for calibration errors, achieving an average absolute trajectory error of 1.09 mm. The reconstructed scenes reach an average target registration error of 2.16 mm, with high visual fidelity (SSIM = 0.69, PSNR = 22.19). These results indicate that our system provides a practical and cost-efficient solution for arthroscopic
navigation, bridging the gap between optical tracking and purely
vision-based systems, and paving the way toward clinically deployable, fully
vision-based arthroscopic guidance.
Read the full abstract ▾
All authors Hongchao Shu, Lalithkumar Seenivasan, Mingxu Liu, Yunseo Hwang, Yu-Chun Ku, Jonathan Knopf, Alejandro Martin-Gomez, Mehran Armand, Mathias Unberath
Systems and Control · Cross-listed from cs.LG
On the Convergence of Overparameterized Problems: Inherent Properties of the Compositional Structure of Neural Networks
Arthur Castello Branco de Oliveira, Dhruv Jatkar and Eduardo Sontag
This paper investigates how the compositional structure of neural networks shapes their optimization landscape and training dynamics. We analyze the gradient flow associated with overparameterized optimization problems, which can be interpreted as training a neural network with linear activations. Remarkably, we show that the global convergence properties can be derived for any cost function that is proper and real analytic. We then specialize the analysis to scalar-valued cost functions, where the geometry of the landscape can be fully characterized. In this setting, we demonstrate that key structural features – such as the location and stability of saddle points – are universal across all admissible costs, depending solely on the overparameterized representation rather than on problem-specific details. Moreover, we show that convergence can be arbitrarily accelerated depending on the initialization, as measured by an imbalance metric introduced in this work. Finally, we discuss how these insights may generalize to neural networks with sigmoidal activations, showing through a simple example which
geometric and dynamical properties persist beyond the linear case.
Read the full abstract ▾
Signal Processing
Investigation of Feature Selection and Pooling Methods for Environmental Sound Classification
Parinaz Binandeh Dehaghani, Danilo Pena and A. Pedro Aguiar
This paper explores the impact of dimensionality reduction and pooling methods for Environmental Sound Classification (ESC) using lightweight CNNs. We evaluate Sparse Salient Region Pooling (SSRP) and its variants, SSRP-Basic (SSRP-B) and SSRP-Top-K (SSRP-T), under various hyperparameter settings and compare them with Principal Component Analysis (PCA). Experiments on the ESC-50
dataset demonstrate that SSRP-T achieves up to 80.69 % accuracy, significantly outperforming both the baseline CNN (66.75 %) and the PCA-reduced model (37.60 %). Our findings confirm that a well-tuned sparse pooling strategy provides a robust, efficient, and high-performing solution for ESC tasks, particularly in resource-constrained scenarios where balancing accuracy and computational cost is crucial.
Read the full abstract ▾
Systems and Control
A Smooth Penalty-Based Feedback Law for Reactive Obstacle Avoidance with Convergence Guarantees
Lyes Smaili and Soulaimane Berkane
This paper addresses the problem of safe autonomous
navigation in unknown obstacle-filled environments using only local sensory information. We propose a smooth feedback controller derived from an unconstrained penalty-based formulation that guarantees safety by construction. The controller modifies an arbitrary nominal input through a closed-form expression. The resulting closed-form feedback has a projection structure that interpolates between the nominal control and its orthogonal projection onto the obstacle boundary, ensuring forward invariance of a user-defined safety margin. The control law depends only on the distance and bearing to obstacles and requires no map, switching, or set construction. When the nominal input is a gradient descent of a
navigation potential, we prove that the closed-loop system achieves almost global asymptotic stability (AGAS) to the goal. Undesired
equilibria are shown to be unstable under a mild
geometric curvature condition, which compares the normal
curvature of the obstacle boundary with that of the potential level sets. We refer to the proposed method as SPF (Safe Penalty-based Feedback), which ensures safe and smooth
navigation with minimal computational overhead, as demonstrated through simulations in complex 2D and 3D environments.
Read the full abstract ▾
Image and Video Processing · Cross-listed from cs.CV
PANDA – Patch And Distribution-Aware Augmentation for Long-Tailed Exemplar-Free Continual Learning
Siddeshwar Raghavan, Jiangpeng He and Fengqing Zhu
Exemplar-Free Continual Learning (EFCL) restricts the storage of previous task data and is highly susceptible to catastrophic forgetting. While pre-trained models (PTMs) are increasingly leveraged for EFCL, existing methods often overlook the inherent imbalance of real-world data distributions. We discovered that real-world data streams commonly exhibit dual-level imbalances,
dataset-level distributions combined with extreme or reversed skews within individual tasks, creating both intra-task and inter-task disparities that hinder effective learning and generalization. To address these challenges, we propose PANDA, a Patch-and-Distribution-Aware Augmentation framework that integrates seamlessly with existing PTM-based EFCL methods. PANDA amplifies low-frequency classes by using a CLIP encoder to identify representative regions and transplanting those into frequent-class samples within each task. Furthermore, PANDA incorporates an adaptive balancing strategy that leverages prior task distributions to smooth inter-task imbalances, reducing the overall gap between average samples across tasks and enabling fairer learning with frozen PTMs. Extensive experiments and ablation studies demonstrate PANDA's capability to work with existing PTM-based CL methods, improving accuracy and reducing catastrophic forgetting.
Read the full abstract ▾
Systems and Control · Cross-listed from cs.RO
A Robust Task-Level Control Architecture for Learned Dynamical Systems
Eshika Pathak, Ahmed Aboudonia, Sandeep Banik et al. (+1 more)
Dynamical system (DS)-based learning from demonstration (LfD) is a powerful tool for generating motion plans in the operation ('task') space of robotic systems. However, realizing generated motion plans is often compromised by a "task-execution mismatch", where unmodeled dynamics, persistent disturbances, and system latency cause the
robot's task-space state to diverge from the desired state. We propose a novel task-level robust control architecture, L1-augmented Dynamical Systems (L1-DS), that explicitly handles the task-execution mismatch in tracking a nominal motion plan generated by any DS-based LfD scheme. Our framework augments any DS-based LfD model with a nominal stabilizing controller and an L1 adaptive controller. Furthermore, we introduce a windowed Dynamic Time Warping (DTW)-based target selector, which enables the nominal stabilizing controller to handle temporal misalignment for improved phase-consistent tracking. We demonstrate the efficacy of our architecture on the LASA and IROS handwriting
datasets.
Read the full abstract ▾
All authors Eshika Pathak, Ahmed Aboudonia, Sandeep Banik, Naira Hovakimyan
Systems and Control
Robust Time-Varying Control Barrier Functions with Sector-Bounded Nonlinearities
Felix Biertümpfel, Jungbae Chun and Peter Seiler
This paper presents a novel approach for ensuring safe operation of systems subject to input nonlinearities and time-varying safety constraints. We extend the time-varying barrier function framework to address time-varying safety constraints and explicitly account for control-dependent nonlinearities at the plant input. Guaranteed bounds on the input-output behavior of these nonlinearities are provided through pointwise-in-time quadratic constraints. The result is a class of robust time-varying control barrier functions that define a safety filter. This filter ensures robust safety for all admissible nonlinearities while minimally modifying the command generated by a baseline controller. We derive a second-order cone program (SOCP) to compute this safety filter online and provide novel feasibility conditions for ball-constrained inputs. The proposed approach is demonstrated on a spacecraft docking maneuver.
Read the full abstract ▾
Signal Processing · Cross-listed from cs.LG
NeuroLingua: A Language-Inspired Hierarchical Framework for Multimodal Sleep Stage Classification Using EEG and EOG
Mahdi Samaee, Mehran Yazdi and Daniel Massicotte
Automated sleep stage classification from polysomnography remains limited by the lack of expressive temporal hierarchies, challenges in
multimodal EEG and EOG fusion, and the limited interpretability of deep learning models. We propose NeuroLingua, a
language-inspired framework that conceptualizes sleep as a structured physiological
language. Each 30-second epoch is decomposed into overlapping 3-second subwindows ("
tokens") using a CNN-based tokenizer, enabling hierarchical temporal modeling through dual-level
Transformers: intra-segment encoding of local dependencies and inter-segment integration across seven consecutive epochs (3.5 minutes) for extended context. Modality-specific
embeddings from
EEG and EOG channels are fused via a Graph Convolutional Network, facilitating robust
multimodal integration. NeuroLingua is evaluated on the Sleep-EDF Expanded and ISRUC-Sleep
datasets, achieving state-of-the-art results on Sleep-EDF (85.3% accuracy, 0.800 macro F1, and 0.796 Cohen's kappa) and competitive performance on ISRUC (81.9% accuracy, 0.802 macro F1, and 0.755 kappa), matching or exceeding published baselines in overall and per-class metrics. The architecture's
attention mechanisms enhance the detection of clinically relevant sleep microevents, providing a principled foundation for future interpretability, explainability, and causal inference in sleep research. By framing sleep as a compositional
language, NeuroLingua unifies hierarchical sequence modeling and
multimodal fusion, advancing automated sleep staging toward more transparent and clinically meaningful applications.
Read the full abstract ▾
Systems and Control · Cross-listed from stat.ME
Modelos Empiricos de Pos-Dupla Selecao por LASSO: Discussoes para Estudos do Transporte Aereo
Alessandro V. M. Oliveira
This paper presents and discusses forms of estimation by regularized regression and model selection using the LASSO method - Least Absolute Shrinkage and Selection Operator. LASSO is recognized as one of the main supervised learning methods applied to high-dimensional econometrics, allowing work with large volumes of data and multiple correlated controls. Conceptual issues related to the consequences of high dimensionality in modern econometrics and the principle of sparsity, which underpins regularization procedures, are addressed. The study examines the main post-double selection and post-regularization models, including variations applied to instrumental variable models. A brief description of the lassopack routine package, its syntaxes, and examples of HD, HDS (High-Dimension Sparse), and IV-HDS models, with combinations involving fixed effects
estimators, is also presented. Finally, the potential application of the approach in research focused on air transport is discussed, with emphasis on an empirical study on the operational efficiency of airlines and aircraft fuel consumption.
Read the full abstract ▾
Signal Processing · Cross-listed from q-bio.NC
Brian Intensify: An Adaptive Machine Learning Framework for Auditory EEG Stimulation and Cognitive Enhancement in FXS
Zag ElSayed, Grace Westerkamp, Jack Yanchen Liu et al. (+1 more)
Neurodevelopmental disorders such as Fragile X Syndrome (FXS) and Autism Spectrum Disorder (ASD) are characterized by disrupted cortical oscillatory activity, particularly in the alpha and gamma frequency bands. These abnormalities are linked to deficits in
attention, sensory processing, and cognitive function. In this work, we present an adaptive machine learning-based
brain-computer interface (BCI) system designed to modulate neural oscillations through frequency-specific auditory stimulation to enhance cognitive readiness in individuals with FXS.
EEG data were recorded from 38 participants using a 128-channel system under a stimulation paradigm consisting of a 30-second baseline (no stimulus) followed by 60-second auditory entrainment episodes at 7Hz, 9Hz, 11Hz, and 13Hz. A comprehensive analysis of power spectral features (Alpha, Gamma, Delta, Theta, Beta) and cross-frequency coupling metrics (Alpha-Gamma, Alpha-Beta, etc.) was conducted. The results identified Peak Alpha Power, Peak Gamma Power, and Alpha Power per second per channel as the most discriminative biomarkers. The 13Hz stimulation condition consistently elicited a significant increase in Alpha activity and suppression of Gamma activity, aligning with our optimization objective. A supervised machine learning framework was developed to predict
EEG responses and dynamically adjust stimulation parameters, enabling real-time, subject-specific adaptation. This work establishes a novel
EEG-driven optimization framework for cognitive neuromodulation, providing a foundational model for next-generation AI-integrated BCI systems aimed at personalized neurorehabilitation in FXS and related disorders.
Read the full abstract ▾
All authors Zag ElSayed, Grace Westerkamp, Jack Yanchen Liu, Ernest Pedapati
Image and Video Processing
A Fourier-Based Global Denoising Model for Smart Artifacts Removing of Microscopy Images
Huanhuan Zhao, Connor Vernachio, Laxmi Bhurtel et al. (+9 more)
Microscopy such as Scanning Tunneling Microscopy (STM), Atomic Force Microscopy (AFM) and Scanning Electron Microscopy (SEM) are essential tools in material imaging at micro- and nanoscale resolutions to extract physical knowledge and materials structure-property relationships. However, tuning microscopy controls (e.g. scanning speed, current setpoint, tip bias etc.) to obtain a high-quality of images is a non-trivial and time-consuming effort. On the other hand, with sub-standard images, the key features are not accurately discovered due to noise and artifacts, leading to erroneous analysis. Existing denoising models mostly build on generalizing the weak signals as noises while the strong signals are enhanced as key features, which is not always the case in microscopy images, thus can completely erase a significant amount of hidden physical information. To address these limitations, we propose a global denoising model (GDM) to smartly remove artifacts of microscopy images while preserving weaker but physically important features. The proposed model is developed based on 1) first designing a two-imaging input channel of non-pair and goal specific pre-processed images with user-defined trade-off information between two channels and 2) then integrating a loss function of pixel- and fast
Fourier-transformed (FFT) based on training the U-net model. We compared the proposed GDM with the non-FFT denoising model over STM-generated images of Copper(Cu) and Silicon(Si) materials, AFM-generated Pantoea sp.YR343 bio-film images and SEM-generated plastic degradation images. We believe this proposed workflow can be extended to improve other microscopy image quality and will benefit the experimentalists with the proposed design flexibility to smartly tune via domain-experts preferences.
Read the full abstract ▾
All authors Huanhuan Zhao, Connor Vernachio, Laxmi Bhurtel, Wooin Yang, Ruben Millan-Solsona, Spenser R. Brown, Marti Checa, Komal Sharma Agrawal, Adam M. Guss, Liam Collins, Wonhee Ko, Arpan Biswas
Systems and Control · Cross-listed from cs.RO
A Shared-Autonomy Construction Robotic System for Overhead Works
David Minkwan Kim, K. M. Brian Lee, Yong Hyeok Seo et al. (+11 more)
We present the ongoing development of a robotic system for overhead work such as ceiling drilling. The hardware platform comprises a mobile base with a two-stage lift, on which a bimanual torso is mounted with a custom-designed drilling end effector and RGB-D cameras. To support teleoperation in dynamic environments with limited visibility, we use Gaussian splatting for online 3D reconstruction and introduce motion parameters to model moving objects. For safe operation around dynamic obstacles, we developed a neural configuration-space barrier approach for planning and control. Initial feasibility studies demonstrate the capability of the hardware in drilling, bolting, and anchoring, and the software in safe teleoperation in a dynamic environment.
Read the full abstract ▾
All authors David Minkwan Kim, K. M. Brian Lee, Yong Hyeok Seo, Nikola Raicevic, Runfa Blark Li, Kehan Long, Chan Seon Yoon, Dong Min Kang, Byeong Jo Lim, Young Pyoung Kim, Nikolay Atanasov, Truong Nguyen, Se Woong Jun, Young Wook Kim
Systems and Control
Statistically Consistent Approximate Model Predictive Control
Elias Milios, Kim P. Wabersich, Felix Berkel et al. (+2 more)
Model Predictive Control (MPC) offers rigorous safety and performance guarantees but is computationally intensive. Approximate MPC (AMPC) aims to circumvent this drawback by learning a computationally cheaper surrogate policy. Common approaches focus on imitation learning (IL) via behavioral cloning (BC), minimizing a mean-squared-error loss on a collection of state-input pairs. However, BC fundamentally fails to provide accurate approximations when MPC solutions are set-valued due to non-convex constraints or local minima. We propose a two-stage IL procedure to accurately approximate nonlinear, potentially set-valued MPC policies. The method integrates an approximation of the MPC's optimal value function into a one-step look-ahead loss function, and thereby embeds the MPC's constraint and performance objectives into the IL objective. This is achieved by adopting a stabilizing soft constrained MPC formulation, which reflects constraint violations in the optimal value function by combining a constraint tightening with slack penalties. We prove statistical consistency for policies that exactly minimize our IL objective, implying convergence to a safe and stabilizing control law, and establish input-to-state stability guarantees for approximate minimizers. Simulations demonstrate improved performance compared to BC.
Read the full abstract ▾
All authors Elias Milios, Kim P. Wabersich, Felix Berkel, Felix Gruber, Melanie N. Zeilinger
Image and Video Processing
TempRetinex: Retinex-based Unsupervised Enhancement for Low-light Video Under Diverse Lighting Conditions
Yini Li, Louis Forster, David Bull et al. (+1 more)
The acquisition of paired low-light
video sequences remains challenging due to issues associated with poor temporal consistency, varying illumination characteristics and camera parameters. This has driven significant interest in unsupervised low-light enhancement approaches. In this context, we propose TempRetinex, an unsupervised Retinex-based
video enhancement framework exploiting inter-frame correlations. We introduce Brightness Consistency Preprocessing (BCP) that explicitly aligns intensity distributions across exposures. BCP is shown to significantly improve model robustness to diverse lighting scenarios. Moreover, we propose a multiscale temporal consistency-aware loss and an occlusion-aware masking technique to enforce similarity between consecutive frames. We further incorporate a Reverse Inference (RI) strategy to refine temporally unstable frames and a Self-Ensemble (SE) mechanism to boost denoising across diverse textures. Experiments demonstrate that TempRetinex achieves state-of-the-art performance in perceptual quality.
Read the full abstract ▾
All authors Yini Li, Louis Forster, David Bull, Nantheera Anantrasirichai
Image and Video Processing
TomoGraphView: 3D Medical Image Classification with Omnidirectional Slice Representations and Graph Neural Networks
Johannes Kiechle, Stefan M. Fischer, Daniel M. Lang et al. (+5 more)
The sharp rise in medical
tomography examinations has created a demand for automated systems that can reliably extract informative features for downstream tasks such as tumor characterization. Although 3D volumes contain richer information than individual slices, effective 3D classification remains difficult: volumetric data encode complex spatial dependencies, and the scarcity of large-scale 3D
datasets has constrained progress toward 3D foundation models. As a result, many recent approaches rely on 2D
vision foundation models trained on natural images, repurposing them as feature extractors for medical scans with surprisingly strong performance. Despite their practical success, current methods that apply 2D foundation models to 3D scans via slice-based decomposition remain fundamentally limited. Standard slicing along axial, sagittal, and coronal planes often fails to capture the true spatial extent of a structure when its orientation does not align with these canonical views. More critically, most approaches aggregate slice features independently, ignoring the underlying 3D geometry and losing spatial coherence across slices. To overcome these limitations, we propose TomoGraphView, a novel framework that integrates omnidirectional volume slicing with spherical graph-based feature aggregation. Instead of restricting the model to axial, sagittal, or coronal planes, our method samples both canonical and non-canonical cross-sections generated from uniformly distributed points on a sphere enclosing the volume. We publicly share our accessible code base at http://github.com/compai-lab/2025-MedIA-kiechle and provide a user-friendly library for omnidirectional volume slicing at https://pypi.org/project/OmniSlicer.
Read the full abstract ▾
All authors Johannes Kiechle, Stefan M. Fischer, Daniel M. Lang, Cosmin I. Bercea, Matthew J. Nyflot, Lina Felsner, Julia A. Schnabel, Jan C. Peeken