arXiv subjects

Ju Liu

Publications and source records attributed to Ju Liu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Lightweight Medical Image Restoration via Integrating Reliable Lesion-Semantic Driven Prior

Medical image restoration tasks aim to recover high-quality images from degraded observations, exhibiting emergent desires in many clinical scenarios, such as low-dose CT image denoising, MRI super-resolution, and MRI artifact removal. Despite the success achieved by existing deep learning-based restoration methods with sophisticated modules, they struggle with rendering computationally-efficient reconstruction results. Moreover, they usually ignore the reliability of the restoration results, which is much more urgent in medical systems. To alleviate these issues, we present LRformer, a Lightweight Transformer-based method via Reliability-guided learning in the frequency domain. Specifically, inspired by the uncertainty quantification in Bayesian neural networks (BNNs), we develop a Reliable Lesion-Semantic Prior Producer (RLPP). RLPP leverages Monte Carlo (MC) estimators with stochastic sampling operations to generate sufficiently-reliable priors by performing multiple inferences on the foundational medical image segmentation model, MedSAM. Additionally, instead of directly incorporating the priors in the spatial domain, we decompose the cross-attention (CA) mechanism into real symmetric and imaginary anti-symmetric parts via fast Fourier transform (FFT), resulting in the design of the Guided Frequency Cross-Attention (GFCA) solver. By leveraging the conjugated symmetric property of FFT, GFCA reduces the computational complexity of naive CA by nearly half. Extensive experimental results in various tasks demonstrate the superiority of the proposed LRformer in both effectiveness and efficiency.

eess.IV↗

arXiv2025

Enhancing Intelligibility for Generative Target Speech Extraction via Joint Optimization with Target Speaker ASR

Target speech extraction (TSE) isolates the speech of a specific speaker from a multi-talker overlapped speech mixture. Most existing TSE models rely on discriminative methods, typically predicting a time-frequency spectrogram mask for the target speech. However, imperfections in these masks often result in over-/under-suppression of target/non-target speech, degrading perceptual quality. Generative methods, by contrast, re-synthesize target speech based on the mixture and target speaker cues, achieving superior perceptual quality. Nevertheless, these methods often overlook speech intelligibility, leading to alterations or loss of semantic content in the re-synthesized speech. Inspired by the Whisper model's success in target speaker ASR, we propose a generative TSE framework based on the pre-trained Whisper model to address the above issues. This framework integrates semantic modeling with flow-based acoustic modeling to achieve both high intelligibility and perceptual quality. Results from multiple benchmarks demonstrate that the proposed method outperforms existing generative and discriminative baselines. We present speech samples on https://aisaka0v0.github.io/GenerativeTSE_demo/.

eess.AS↗

arXiv2025

CLAPSep: Leveraging Contrastive Pre-trained Model for Multi-Modal Query-Conditioned Target Sound Extraction

Universal sound separation (USS) aims to extract arbitrary types of sounds from real-world recordings. This can be achieved by language-queried target sound extraction (TSE), which typically consists of two components: a query network that converts user queries into conditional embeddings, and a separation network that extracts the target sound accordingly. Existing methods commonly train models from scratch. As a consequence, substantial data and computational resources are required to make the randomly initialized model comprehend sound events and perform separation accordingly. In this paper, we propose to integrate pre-trained models into TSE models to address the above issue. To be specific, we tailor and adapt the powerful contrastive language-audio pre-trained model (CLAP) for USS, denoted as CLAPSep. CLAPSep also accepts flexible user inputs, taking both positive and negative user prompts of uni- and/or multi-modalities for target sound extraction. These key features of CLAPSep can not only enhance the extraction performance but also improve the versatility of its application. We provide extensive experiments on 5 diverse datasets to demonstrate the superior performance and zero- and few-shot generalizability of our proposed CLAPSep with fast training convergence, surpassing previous methods by a significant margin. Full codes and some audio examples are released for reproduction and evaluation.

eess.AS↗

arXiv2025

Language-Queried Target Sound Extraction Without Parallel Training Data

Language-queried target sound extraction (TSE) aims to extract specific sounds from mixtures based on language queries. Traditional fully-supervised training schemes require extensively annotated parallel audio-text data, which are labor-intensive. We introduce a parallel-data-free training scheme, requiring only unlabelled audio clips for TSE model training by utilizing the contrastive language-audio pre-trained model (CLAP). In a vanilla parallel-data-free training stage, target audio is encoded using the pre-trained CLAP audio encoder to form a condition embedding, while during testing, user language queries are encoded by CLAP text encoder as the condition embedding. This vanilla approach assumes perfect alignment between text and audio embeddings, which is unrealistic. Two major challenges arise from training-testing mismatch: the persistent modality gap between text and audio and the risk of overfitting due to the exposure of rich acoustic details in target audio embedding during training. To address this, we propose a retrieval-augmented strategy. Specifically, we create an embedding cache using audio captions generated by a large language model (LLM). During training, target audio embeddings retrieve text embeddings from this cache to use as condition embeddings, ensuring consistent modalities between training and testing and eliminating information leakage. Extensive experiment results show that our retrieval-augmented approach achieves consistent and notable performance improvements over existing state-of-the-art with better generalizability.

eess.AS↗

arXiv2025

Modeling finite viscoelasticity based on the Green-Naghdi kinematic assumption and generalized strains

We propose a modeling framework for finite viscoelasticity, inspired by the kinematic assumption made by Green and Naghdi in plasticity. This approach fundamentally differs from the widely used multiplicative decomposition of the deformation gradient, as the intermediate configuration, a concept that remains debated, becomes unnecessary. The advent of the concept of generalized strains allows the Green-Naghdi assumption to be employed with different strains, offering a flexible mechanism to separate inelastic deformation from total deformation. This leads to a constitutive theory in which the kinematic separation is adjustable and can be calibrated. For quadratic configurational free energy, the framework yields a suite of finite linear viscoelasticity models governed by linear evolution equations. Notably, these models recover established models, including those by Green and Tobolsky (1946) and Simo (1987), when the Seth-Hill strain is chosen with the strain parameter being -2 and 2, respectively. It is also related to the model of Miehe and Keck (2000) when the strain is of the Hencky type. We further extend the approach by adopting coercive strains, which allows us to define an elastic deformation tensor locally. This facilitates modeling the viscous branch using general forms of the configurational free energy, and we construct a micromechanical viscoelastic model as a representative instantiation. The constitutive integration algorithms of the proposed models are detailed. We employ the experimental data of VHB 4910 to examine the proposed models, which demonstrate their effectiveness and potential advantages in the quality of fitting and prediction. Three-dimensional finite element analysis is also conducted to assess the influence of different strains on the viscoelastic behavior.

math.NA↗

arXiv2024

An Efficient Sum-Rate Maximization Algorithm for Fluid Antenna-Assisted ISAC System

This letter investigates a fluid antenna (FA)-assisted integrated sensing and communication (ISAC) system, with joint antenna position optimization and waveform design. We consider enhancing the sum-rate maximization (SRM) and sensing performance with the aid of FAs. Although the introduction of FAs brings more degrees of freedom for performance optimization, its position optimization poses a non-convex programming problem and brings great computational challenges. This letter contributes to building an efficient design algorithm by the block successive upper bound minimization and majorization-minimization principles, with each step admitting closed-form update for the ISAC waveform design. In addition, the extrapolation technique is exploited further to speed up the empirical convergence of FA position design. Simulation results show that the proposed design can achieve state-of-the-art sum-rate performance with at least 60% computation cutoff compared to existing works with successive convex approximation (SCA) and particle swarm optimization (PSO) algorithms.

cs.IT↗

arXiv2024

MVPatch: More Vivid Patch for Adversarial Camouflaged Attacks on Object Detectors in the Physical World

Recent studies have shown that Adversarial Patches (APs) can effectively manipulate object detection models. However, the conspicuous patterns often associated with these patches tend to attract human attention, posing a significant challenge. Existing research has primarily focused on enhancing attack efficacy in the physical domain while often neglecting the optimization of stealthiness and transferability. Furthermore, applying APs in real-world scenarios faces major challenges related to transferability, stealthiness, and practicality. To address these challenges, we introduce generalization theory into the context of APs, enabling our iterative process to simultaneously enhance transferability and refine visual correlation with realistic images. We propose a Dual-Perception-Based Framework (DPBF) to generate the More Vivid Patch (MVPatch), which enhances transferability, stealthiness, and practicality. The DPBF integrates two key components: the Model-Perception-Based Module (MPBM) and the Human-Perception-Based Module (HPBM), along with regularization terms. The MPBM employs ensemble strategy to reduce object confidence scores across multiple detectors, thereby improving AP transferability with robust theoretical support. Concurrently, the HPBM introduces a lightweight method for achieving visual similarity, creating natural and inconspicuous adversarial patches without relying on additional generative models. The regularization terms further enhance the practicality of the generated APs in the physical domain. Additionally, we introduce naturalness and transferability scores to provide an unbiased assessment of APs. Extensive experimental validation demonstrates that MVPatch achieves superior transferability and a natural appearance in both digital and physical domains, underscoring its effectiveness and stealthiness.

cs.CR↗

arXiv2024

Modeling fibrous tissue in vascular fluid-structure interaction: a morphology-based pipeline and biomechanical significance

We propose a suite of technologies for analyzing the interaction between anisotropic arterial walls and blood flow for subject-specific geometries. Utilizing an established lumen modeling strategy, we present a comprehensive pipeline for generating the thick-walled artery models. Through a specialized mesh generation procedure, we obtain the meshes for the arterial lumen and wall with mesh continuity across the interface ensured. Exploiting the centerline information, a series of procedures is introduced for generating local basis vectors within the arterial wall. The procedures are tailored to handle thick-walled and, in particular, aneurysmatic tissues in which the basis vectors may exhibit transmural variations. Additionally, we propose methods to accurately identify the centerline in multi-branched vessels and bifurcating regions. The developed fiber generation method is evaluated against the strategy using linear elastic analysis, demonstrating that the proposed approach yields satisfactory fiber definitions in the considered benchmark. Finally, we examine the impact of anisotropic arterial wall models on the vascular fluid-structure interaction analysis through numerical examples. For comparison purposes, the neo-Hookean model is considered. The first case involves an idealized curved geometry, while the second case studies an image-based abdominal aorta model. The numerical results reveal that the deformation and stress distribution are critically related to the constitutive model of the wall, while the hemodynamic factors are less sensitive to the wall model. This work paves the way for more accurate image-based vascular modeling and enhances the prediction of arterial behavior under physiologically realistic conditions.

physics.med-ph↗

arXiv2024

A continuum and computational framework for viscoelastodynamics: III. A nonlinear theory

We continue our investigation of viscoelasticity by extending the Holzapfel-Simo approach discussed in Part I to the fully nonlinear regime. By scrutinizing the relaxation property for the non-equilibrium stresses, it is revealed that a kinematic assumption akin to the Green-Naghdi type is necessary in the design of the potential. This insight underscores a link between the so-called additive plasticity and the viscoelasticity model under consideration, further inspiring our development of a nonlinear viscoelasticity theory. Our strategy is based on Hill's hyperelasticity framework and leverages the concept of generalized strains. Notably, the adopted kinematic assumption makes the proposed theory fundamentally different from the existing models rooted in the notion of the intermediate configuration. The computation aspects, including the consistent linearization, constitutive integration, and modular implementation, are addressed in detail. A suite of numerical examples is provided to demonstrate the capability of the proposed model in characterizing viscoelastic material behaviors at large strains.

math.NA↗

arXiv2024

An efficient algorithm for multiuser sum-rate maximization of large-scale active RIS-aided MIMO system

Active reconfigurable intelligent surface (RIS) is a new RIS architecture that can reflect and amplify communication signals. It can provide enhanced performance gain compared to the conventional passive RIS systems that can only reflect the signals. On the other hand, the design problem of active RIS-aided systems is more challenging than the passive RIS-aided systems and its efficient algorithms are less studied. In this paper, we consider the sum rate maximization problem in the multiuser massive multiple-input single-output (MISO) downlink with the aid of a large-scale active RIS. Existing approaches for handling this problem usually resort to general optimization solvers and can be computationally prohibitive. We propose an efficient block successive upper bound minimization (BSUM) method, of which each step has a (semi) closed-form update. Thus, the proposed algorithm has an attractive low per-iteration complexity. By simulation, our proposed algorithm consumes much less computation than the existing approaches. In particular, when the MIMO and/or RIS sizes are large, our proposed algorithm can be orders-of-magnitude faster than existing approaches.

cs.IT↗

arXiv2024

Extending Whisper with prompt tuning to target-speaker ASR

Target-speaker automatic speech recognition (ASR) aims to transcribe the desired speech of a target speaker from multi-talker overlapped utterances. Most of the existing target-speaker ASR (TS-ASR) methods involve either training from scratch or fully fine-tuning a pre-trained model, leading to significant training costs and becoming inapplicable to large foundation models. This work leverages prompt tuning, a parameter-efficient fine-tuning approach, to extend Whisper, a large-scale single-talker ASR model, to TS-ASR. Variants of prompt tuning approaches along with their configurations are explored and optimized for TS-ASR.Experimental results show that prompt tuning can achieve performance comparable to state-of-the-art full training approaches while only requiring about 1\% of task-specific model parameters. Notably, the original Whisper's features, such as inverse text normalization and timestamp tagging, are retained in target-speaker ASR, keeping the generated transcriptions natural and informative.

cs.CL↗

arXiv2023

A continuum and computational framework for viscoelastodynamics: II. Strain-driven and energy-momentum consistent schemes

We continue our investigation of finite deformation linear viscoelastodynamics by focusing on constructing accurate and reliable numerical schemes. The concrete thermomechanical foundation developed in the previous study paves the way for pursuing discrete formulations with critical physical and mathematical structures preserved. Energy stability, momentum conservation, and temporal accuracy constitute the primary factors in our algorithm design. For inelastic materials, the directionality condition, a property for the stress to be energy consistent, is extended with the dissipation effect taken into account. Moreover, the integration of the constitutive relations calls for an algorithm design of the internal state variables and their conjugate variables. A directionality condition for the conjugate variables is introduced as an indispensable ingredient for ensuring physically correct numerical dissipation. By leveraging the particular structure of the configurational free energy, a set of update formulas for the internal state variables is obtained. Detailed analysis reveals that the overall discrete schemes are energy-momentum consistent and achieve first- and second-order accuracy in time, respectively. Numerical examples are provided to justify the appealing features of the proposed methodology.

math.NA↗

arXiv2023

A structure-preserving integrator for incompressible finite elastodynamics based on a grad-div stabilized mixed formulation with particular emphasis on stretch-based material models

We present a structure-preserving scheme based on a recently-proposed mixed formulation for incompressible hyperelasticity formulated in principal stretches. Although there exist Hamiltonians introduced for quasi-incompressible elastodynamics based on different variational formulations, the one in the fully incompressible regime has yet been identified in the literature. The adopted mixed formulation naturally provides a new Hamiltonian for fully incompressible elastodynamics. Invoking the discrete gradient formula, we are able to design fully-discrete schemes that preserve the Hamiltonian and momenta. The scaled mid-point formula, another popular option for constructing algorithmic stresses, is analyzed and demonstrated to be non-robust numerically. The generalized Taylor-Hood element based on the spline technology conveniently provides a higher-order, robust, and inf-sup stable spatial discretization option for finite strain analysis. To enhance the element performance in volume conservation, the grad-div stabilization, a technique initially developed in computational fluid dynamics, is introduced here for elastodynamics. It is shown that the stabilization term does not impose additional restrictions for the algorithmic stress to respect the invariants, leading to an energy-decaying and momentum-conserving fully discrete scheme. A set of numerical examples is provided to justify the claimed properties. The grad-div stabilization is found to enhance the discrete mass conservation effectively. Furthermore, in contrast to conventional algorithms based on Cardano's formula and perturbation techniques, the spectral decomposition algorithm developed by Scherzinger and Dohrmann is robust and accurate to ensure the discrete conservation laws and is thus recommended for stretch-based material modeling.

math.NA↗

arXiv2022

On the design of energy-decaying momentum-conserving integrator for nonlinear dynamics using energy splitting and perturbation techniques

This work proposes a suite of numerical techniques to facilitate the design of structure-preserving integrators for nonlinear dynamics. The celebrated LaBudde-Greenspan integrator and various energy-momentum schemes adopt a difference quotient formula in their algorithmic force definitions, which suffers from numerical instability as the denominator gets close to zero. There is a need to develop structure-preserving integrators without invoking the quotient formula. In this work, the potential energy of a Hamiltonian system is split into two parts, and specially developed quadrature rules are applied separately to them. The resulting integrators can be regarded as classical ones perturbed with first- or second-order terms, and the energy split guarantees the dissipative nature in the numerical residual. In the meantime, the conservation of invariants is respected in the design. A complete analysis of the proposed integrators is given, with representative numerical examples provided to demonstrate their performance. They can be used either independently as energy-decaying and momentum-conserving schemes for nonlinear problems or as an alternate option with a conserving integrator, such as the LaBudde-Greenspan integrator, when the numerical instability in the difference quotient is detected.

math.NA↗

arXiv2022

Nonlinear Kernel Support Vector Machine with 0-1 Soft Margin Loss

Recent advance on linear support vector machine with the 0-1 soft margin loss ($L_{0/1}$-SVM) shows that the 0-1 loss problem can be solved directly. However, its theoretical and algorithmic requirements restrict us extending the linear solving framework to its nonlinear kernel form directly, the absence of explicit expression of Lagrangian dual function of $L_{0/1}$-SVM is one big deficiency among of them. In this paper, by applying the nonparametric representation theorem, we propose a nonlinear model for support vector machine with 0-1 soft margin loss, called $L_{0/1}$-KSVM, which cunningly involves the kernel technique into it and more importantly, follows the success on systematically solving its linear task. Its optimal condition is explored theoretically and a working set selection alternating direction method of multipliers (ADMM) algorithm is introduced to acquire its numerical solution. Moreover, we firstly present a closed-form definition to the support vector (SV) of $L_{0/1}$-KSVM. Theoretically, we prove that all SVs of $L_{0/1}$-KSVM are only located on the parallel decision surfaces. The experiment part also shows that $L_{0/1}$-KSVM has much fewer SVs, simultaneously with a decent predicting accuracy, when comparing to its linear peer $L_{0/1}$-SVM and the other six nonlinear benchmark SVM classifiers.

cs.LG↗

arXiv2022

Vascular fluid-structure interaction: unified continuum formulation, image-based mesh generation pipeline, and scalable fully implicit solver technology

We propose a computational framework for vascular fluid-structure interaction (FSI), focusing on biomechanical modeling, geometric modeling, and solver technology. The biomechanical model is constructed based on the unified continuum formulation. We highlight that the chosen time integration scheme differs from existing implicit FSI integration methods in that it is indeed second-order accurate, does not suffer from the overshoot phenomenon, and optimally dissipates high-frequency modes in both subproblems. We propose a pipeline for generating subject-specific meshes for FSI analysis for anatomically realistic geometric modeling. Unlike most existing methodologies that operate directly on the wall surface mesh, our pipeline starts from the image segmentation stage. With high-quality surface meshes obtained, the volumetric meshes are then generated, guaranteeing a boundary-layered mesh in the fluid subdomain and a matching mesh across the fluid-solid interface. In the last, we propose a combined suite of nonlinear and linear solver technologies. Invoking a segregated algorithm within the Newton-Raphson iteration, the problem reduces to solving two linear systems in the multi-corrector stage. The first linear system can be addressed by the algebraic multigrid (AMG) method. The matrix related to the balance equations presents a two-by-two block structure in both subproblems. Using the Schur complement reduction (SCR) technique reduces the problem to solving matrices of smaller sizes of the elliptic type, and the AMG method again becomes a natural candidate. The benefit of the unified formulation is demonstrated in parallelizing the solution algorithms as the number of unknowns matches in both subdomains. We use the Greenshields-Weller benchmark as well as a patient-specific vascular model to demonstrate the robustness, efficiency, and scalability of the overall FSI solver technology.

physics.flu-dyn↗

arXiv2022

Validation of the Reduced Unified Continuum Formulation Against In Vitro 4D-Flow MRI

In our recent work, we introduced the reduced unified continuum formulation for vascular fluid-structure interaction (FSI) and demonstrated enhanced solver accuracy, scalability, and performance compared to conventional approaches. We further verified the formulation against Womersley's deformable wall theory. In this study, we assessed its performance in a compliant patient-specific aortic model by leveraging 3D printing, 2D magnetic resonance imaging (MRI), and 4D-flow MRI to extract high-resolution anatomical and hemodynamic information from an in vitro flow circuit. To accurately reflect experimental conditions, we additionally enabled in-plane vascular motion at each inlet and outlet, and implemented viscoelastic external tissue support and vascular tissue prestressing. Validation of our formulation is achieved through close quantitative agreement in pressures, lumen area changes, pulse wave velocity, and early systolic velocities, as well as qualitative agreement in late systolic flow structures. Our validated suite of FSI techniques can be used to investigate vascular disease initiation, progression, and treatment at a computational cost on the same order as that of rigid-walled simulations. This study is the first to validate a cardiovascular FSI formulation against an in vitro flow circuit involving a compliant vascular phantom of complex patient-specific anatomy.

physics.comp-ph↗

arXiv2022

Numerical investigation of abdominal aortic aneurysm hemodynamics using the reduced unified continuum formulation for vascular fluid-structure interaction

We recently demonstrated the reduction of the unified continuum and variational multiscale formulation to a computationally efficient fluid-structure interaction (FSI) formulation via three sound modeling assumptions pertaining to the vascular wall. Similar to the coupled momentum method introduced by Figueroa et al., the resulting semi-discrete formulation yields a monolithically coupled FSI system posed in an Eulerian frame of reference with only a minor modification of the fluid boundary integral. To achieve uniform second-order temporal accuracy and user-controlled high-frequency algorithmic damping, we adopt the generalized-$α$ method for uniform temporal discretization of the entire coupled system. In conjunction with a fully consistent, segregated predictor multi-corrector algorithm preserving the block structure of the incompressible Navier-Stokes equations in the implicit solver's associated linear system, a three-level nested block preconditioner is adopted for improved representation of the Schur complement. In this work, we apply our reduced unified continuum formulation to an appropriately prestressed patient-specific abdominal aortic aneurysm and investigate the effects of varying spatial distributions of wall properties on hemodynamic and vascular wall quantities of interest.

physics.med-ph↗