arXiv subjects

Pengfei Zhu

Publications and source records attributed to Pengfei Zhu.

At least 55 records · Page 3Linked to original sources

arXiv2026

Non-Invasive Diagnosis for Clubroot Using Terahertz Time-Domain Spectroscopy and Physics-Constrained Neural Networks

Clubroot, a major soilborne disease affecting canola and other cruciferous crops, is characterized by the development of large galls on the roots of susceptible hosts. In this study, we present the first application of terahertz time-domain spectroscopy (THz-TDS) as a non-invasive diagnosis tool in plant pathology. Compared with conventional molecular, spectroscopic, and immunoassay-based methods, THz-TDS offers distinct advantages, including non-contact, non-destructive, and preparation-free measurement, enabling rapid in situ screening of plant and soil samples. Our results demonstrate that THz-TDS can differentiate between healthy and clubroot-infected tissues by detecting both structural and biochemical alterations. Specifically, infected roots exhibit a blue shift in the refractive index in the low-frequency THz range, along with distinct peaks-indicative of disruptions in water transport and altered metabolic activity in both roots and leaves. Interestingly, the characteristic root swelling observed in infected plants reflects internal tissue disorganization rather than an actual increase in water content. Furthermore, a physics-constrained neural network is proposed to extract the main feature in THz-TDS. A comprehensive evaluation, including time-domain signals, amplitude and phase images, refractive index and absorption coefficient maps, and principal component analysis, provides enhanced contrast and spatial resolution compared to raw time-domain or frequency signals. These findings suggest that THz-TDS holds significant potential for early, non-destructive detection of plant diseases and may serve as a valuable tool to limit their spread in agricultural systems.

eess.IV↗

arXiv2026

CtrlFuse: Mask-Prompt Guided Controllable Infrared and Visible Image Fusion

Infrared and visible image fusion generates all-weather perception-capable images by combining complementary modalities, enhancing environmental awareness for intelligent unmanned systems. Existing methods either focus on pixel-level fusion while overlooking downstream task adaptability or implicitly learn rigid semantics through cascaded detection/segmentation models, unable to interactively address diverse semantic target perception needs. We propose CtrlFuse, a controllable image fusion framework that enables interactive dynamic fusion guided by mask prompts. The model integrates a multi-modal feature extractor, a reference prompt encoder (RPE), and a prompt-semantic fusion module (PSFM). The RPE dynamically encodes task-specific semantic prompts by fine-tuning pre-trained segmentation models with input mask guidance, while the PSFM explicitly injects these semantics into fusion features. Through synergistic optimization of parallel segmentation and fusion branches, our method achieves mutual enhancement between task performance and fusion quality. Experiments demonstrate state-of-the-art results in both fusion controllability and segmentation accuracy, with the adapted task branch even outperforming the original segmentation model.

cs.CV↗

arXiv2026

DIFF-MF: A Difference-Driven Channel-Spatial State Space Model for Multi-Modal Image Fusion

Multi-modal image fusion aims to integrate complementary information from multiple source images to produce high-quality fused images with enriched content. Although existing approaches based on state space model have achieved satisfied performance with high computational efficiency, they tend to either over-prioritize infrared intensity at the cost of visible details, or conversely, preserve visible structure while diminishing thermal target salience. To overcome these challenges, we propose DIFF-MF, a novel difference-driven channel-spatial state space model for multi-modal image fusion. Our approach leverages feature discrepancy maps between modalities to guide feature extraction, followed by a fusion process across both channel and spatial dimensions. In the channel dimension, a channel-exchange module enhances channel-wise interaction through cross-attention dual state space modeling, enabling adaptive feature reweighting. In the spatial dimension, a spatial-exchange module employs cross-modal state space scanning to achieve comprehensive spatial fusion. By efficiently capturing global dependencies while maintaining linear computational complexity, DIFF-MF effectively integrates complementary multi-modal features. Experimental results on the driving scenarios and low-altitude UAV datasets demonstrate that our method outperforms existing approaches in both visual quality and quantitative evaluation.

cs.CV↗

arXiv2026

Towards precision quantitative measurement of radiation reaction within the classical radiation-dominated regime

Radiation reaction (RR) is a fundamental yet incompletely validated process in laser-particle interactions, since it lacks quantitatively definitive experimental verifications, especially the transition from classical to quantum regime. Herein, we propose a novel experimental scenario for investigating radiation RR within the classical radiation-dominated regime (CRDR), via the collision of a high-intensity petawatt-class laser with a tens-of-MeV electron beam from a LINAC. This approach enables access to a distinct parameter regime wherein RR dominates electron dynamics while quantum effects remain modest. Numerical simulations demonstrate that three key observables exist for identifying the RR within this CRDR regime: (i) quantitative measurement of energy spectra to validate the quantum correction factor; (ii) control of the collision time delay with charge-counting to map intensity dependence of RR; and (iii) verification of large angle ($90^\circ$) photon emission under the recoil condition $2γ\gtrsim a_0$. These experimental measurements will establish the benchmarks for RR models spanning the classical-to-quantum regime, thereby providing critical insights into fundamental strong-field quantum electrodynamics.

physics.optics↗

arXiv2025

Improving Partially Observed Trajectories Forecasting by Target-driven Self-Distillation

Accurate prediction of future trajectories of traffic agents is essential for ensuring safe autonomous driving. However, partially observed trajectories can significantly degrade the performance of even state-of-the-art models. Previous approaches often rely on knowledge distillation to transfer features from fully observed trajectories to partially observed ones. This involves firstly training a fully observed model and then using a distillation process to create the final model. While effective, they require multi-stage training, making the training process very expensive. Moreover, knowledge distillation can lead to a performance degradation of the model. In this paper, we introduce a Target-drivenSelf-Distillation method (TSD) for motion forecasting. Our method leverages predicted accurate targets to guide the model in making predictions under partial observation conditions. By employing self-distillation, the model learns from the feature distributions of both fully observed and partially observed trajectories during a single end-to-end training process. This enhances the model's ability to predict motion accurately in both fully observed and partially observed scenarios. We evaluate our method on multiple datasets and state-of-the-art motion forecasting models. Extensive experimental results demonstrate that our approach achieves significant performance improvements in both settings. To facilitate further research, we will release our code and model checkpoints.

cs.CV↗

arXiv2025

Dream-IF: Dynamic Relative EnhAnceMent for Image Fusion

Image fusion aims to integrate comprehensive information from images acquired through multiple sources. However, images captured by diverse sensors often encounter various degradations that can negatively affect fusion quality. Traditional fusion methods generally treat image enhancement and fusion as separate processes, overlooking the inherent correlation between them; notably, the dominant regions in one modality of a fused image often indicate areas where the other modality might benefit from enhancement. Inspired by this observation, we introduce the concept of dominant regions for image enhancement and present a Dynamic Relative EnhAnceMent framework for Image Fusion (Dream-IF). This framework quantifies the relative dominance of each modality across different layers and leverages this information to facilitate reciprocal cross-modal enhancement. By integrating the relative dominance derived from image fusion, our approach supports not only image restoration but also a broader range of image enhancement applications. Furthermore, we employ prompt-based encoding to capture degradation-specific details, which dynamically steer the restoration process and promote coordinated enhancement in both multi-modal image fusion and image enhancement scenarios. Extensive experimental results demonstrate that Dream-IF consistently outperforms its counterparts. The code is publicly available.\footnote{ https://github.com/jehovahxu/Dream-IF

cs.CV↗

arXiv2025

MatMart: Material Reconstruction of 3D Objects via Diffusion

Applying diffusion models to physically-based material estimation and generation has recently gained prominence. In this paper, we propose \ttt, a novel material reconstruction framework for 3D objects, offering the following advantages. First, \ttt\ adopts a two-stage reconstruction, starting with accurate material prediction from inputs and followed by prior-guided material generation for unobserved views, yielding high-fidelity results. Second, by utilizing progressive inference alongside the proposed view-material cross-attention (VMCA), \ttt\ enables reconstruction from an arbitrary number of input images, demonstrating strong scalability and flexibility. Finally, \ttt\ achieves both material prediction and generation capabilities through end-to-end optimization of a single diffusion model, without relying on additional pre-trained models, thereby exhibiting enhanced stability across various types of objects. Extensive experiments demonstrate that \ttt\ achieves superior performance in material reconstruction compared to existing methods.

cs.GR↗

arXiv2025

Point Cloud Quantization through Multimodal Prompting for 3D Understanding

Vector quantization has emerged as a powerful tool in large-scale multimodal models, unifying heterogeneous representations through discrete token encoding. However, its effectiveness hinges on robust codebook design. Current prototype-based approaches relying on trainable vectors or clustered centroids fall short in representativeness and interpretability, even as multimodal alignment demonstrates its promise in vision-language models. To address these limitations, we propose a simple multimodal prompting-driven quantization framework for point cloud analysis. Our methodology is built upon two core insights: 1) Text embeddings from pre-trained models inherently encode visual semantics through many-to-one contrastive alignment, naturally serving as robust prototype priors; and 2) Multimodal prompts enable adaptive refinement of these prototypes, effectively mitigating vision-language semantic gaps. The framework introduces a dual-constrained quantization space, enforced by compactness and separation regularization, which seamlessly integrates visual and prototype features, resulting in hybrid representations that jointly encode geometric and semantic information. Furthermore, we employ Gumbel-Softmax relaxation to achieve differentiable discretization while maintaining quantization sparsity. Extensive experiments on the ModelNet40 and ScanObjectNN datasets clearly demonstrate the superior effectiveness of the proposed method.

cs.CV↗

arXiv2025

BasicAVSR: Arbitrary-Scale Video Super-Resolution via Image Priors and Enhanced Motion Compensation

Arbitrary-scale video super-resolution (AVSR) aims to enhance the resolution of video frames, potentially at various scaling factors, which presents several challenges regarding spatial detail reproduction, temporal consistency, and computational complexity. In this paper, we propose a strong baseline BasicAVSR for AVSR by integrating four key components: 1) adaptive multi-scale frequency priors generated from image Laplacian pyramids, 2) a flow-guided propagation unit to aggregate spatiotemporal information from adjacent frames, 3) a second-order motion compensation unit for more accurate spatial alignment of adjacent frames, and 4) a hyper-upsampling unit to generate scale-aware and content-independent upsampling kernels. To meet diverse application demands, we instantiate three propagation variants: (i) a unidirectional RNN unit for strictly online inference, (ii) a unidirectional RNN unit empowered with a limited lookahead that tolerates a small output delay, and (iii) a bidirectional RNN unit designed for offline tasks where computational resources are less constrained. Experimental results demonstrate the effectiveness and adaptability of our model across these different scenarios. Through extensive experiments, we show that BasicAVSR significantly outperforms existing methods in terms of super-resolution quality, generalization ability, and inference speed. Our work not only advances the state-of-the-art in AVSR but also extends its core components to multiple frameworks for diverse scenarios. The code is available at https://github.com/shangwei5/BasicAVSR.

cs.CV↗

arXiv2025

Decoupled Multi-Predictor Optimization for Inference-Efficient Model Tuning

Recently, remarkable progress has been made in large-scale pre-trained model tuning, and inference efficiency is becoming more crucial for practical deployment. Early exiting in conjunction with multi-stage predictors, when cooperated with a parameter-efficient fine-tuning strategy, offers a straightforward way to achieve an inference-efficient model. However, a key challenge remains unresolved: How can early stages provide low-level fundamental features to deep stages while simultaneously supplying high-level discriminative features to early-stage predictors? To address this problem, we propose a Decoupled Multi-Predictor Optimization (DMPO) method to effectively decouple the low-level representative ability and high-level discriminative ability in early stages. First, in terms of architecture, we introduce a lightweight bypass module into multi-stage predictors for functional decomposition of shallow features from early stages, while a high-order statistics-based predictor is developed for early stages to effectively enhance their discriminative ability. To reasonably train our multi-predictor architecture, a decoupled optimization is proposed to allocate two-phase loss weights for multi-stage predictors during model tuning, where the initial training phase enables the model to prioritize the acquisition of discriminative ability of deep stages via emphasizing representative ability of early stages, and the latter training phase drives discriminative ability towards earlier stages as much as possible. As such, our DMPO can effectively decouple representative and discriminative abilities in early stages in terms of architecture design and model optimization. Experiments across various datasets and pre-trained backbones demonstrate that DMPO clearly outperforms its counterparts when reducing computational cost.

cs.CV↗

arXiv2025

Lithium Niobate Vertical Cavity Electro-Optic Modulator

Electro-optic modulators (EOMs) are vital for optical imaging and information processing, with free-space devices enabling LiDAR and beam control. Lithium niobate (LN), powered by the strong Pockels effect and scalable LN-on-insulator (LNOI) platform, has become a leading material for high-performance EOMs. Here we realize a vertical-cavity EOM in which an LN membrane is sandwiched between two photonic crystal (PhC) mirrors with integrated electrodes. The cavity supports sharp defect-mode resonances that shift efficiently under the Pockels effect, enabling strong modulation of transmission. Experiments show a depth of 43 % at 50 V and a bandwidth of 5 MHz. This architecture combines free-space compatibility with fabrication simplicity, opening new routes to compact electro-optic platforms for ranging, holography, and beam steering.

physics.optics↗

arXiv2025

Multimodal Negative Learning

Multimodal learning systems often encounter challenges related to modality imbalance, where a dominant modality may overshadow others, thereby hindering the learning of weak modalities. Conventional approaches often force weak modalities to align with dominant ones in "Learning to be (the same)" (Positive Learning), which risks suppressing the unique information inherent in the weak modalities. To address this challenge, we offer a new learning paradigm: "Learning Not to be" (Negative Learning). Instead of enhancing weak modalities' target-class predictions, the dominant modalities dynamically guide the weak modality to suppress non-target classes. This stabilizes the decision space and preserves modality-specific information, allowing weak modalities to preserve unique information without being over-aligned. We proceed to reveal multimodal learning from a robustness perspective and theoretically derive the Multimodal Negative Learning (MNL) framework, which introduces a dynamic guidance mechanism tailored for negative learning. Our method provably tightens the robustness lower bound of multimodal learning by increasing the Unimodal Confidence Margin (UCoM) and reduces the empirical error of weak modalities, particularly under noisy and imbalanced scenarios. Extensive experiments across multiple benchmarks demonstrate the effectiveness and generalizability of our approach against competing methods. The code will be available at https://github.com/BaoquanGong/Multimodal-Negative-Learning.git.

cs.LG↗

arXiv2025

A review of cultural heritage inspection: Toward terahertz from mid-infrared region

This review explores non-invasive imaging (NII) methods covering the mid- and far-infrared to the terahertz spectral regions (up to approximately 1000 um) for the detection and analysis of cultural heritage artifacts. In the thermal infrared domain, where radiation follows Planck's law, the self-emission of materials reveals intrinsic properties and internal degradation. By contrast, in the near-infrared range, external illumination enhances surface details and pigment differentiation. Far-infrared and terahertz techniques, operating in both transmission and reflection modes, provide complementary insights by penetrating surface layers to uncover subsurface structures and concealed features. Integrating visible and infrared imaging further enriches diagnostic capabilities by correlating conventional visual assessments with spectral information. Beyond reviewing the wide applications of these NII techniques in cultural heritage research, this work also summarizes recent advances in signal processing, encompassing both hardware and software developments. In particular, deep learning has revolutionized the field by enabling automated classification, feature extraction, defect detection, and super-resolution imaging. Through supervised and unsupervised learning strategies, neural networks can reliably identify subtle anomalies and material variations indicative of past restorations or early stages of deterioration. In conclusion, the convergence of advanced spectral imaging, sophisticated signal processing, and deep neural networks offers a transformative pathway toward more accurate, efficient, and data-driven cultural heritage analysis, ultimately supporting more informed conservation and restoration decisions.

physics.optics↗

arXiv2025

THz-PINNs: Time-Domain Forward Modeling of Terahertz Spectroscopy with Physics-Informed Neural Networks

Terahertz time-domain spectroscopy (THz-TDS) is a powerful tool for extracting optical and electrical parameters, as well as probing internal structures and surface morphology of materials. However, conventional simulation techniques, such as the finite element method (FEM) and finite-difference time-domain (FDTD), face challenges in accurately modeling THz wave propagation. These difficulties arise from the broad operating bandwidth of THz-TDS (~0.1-10 THz), which demands extremely high temporal resolution to capture pulse dynamics and fine spatial resolution to resolve microstructures and interface effects. To address these limitations, we introduce physics-informed neural networks (PINNs) into THz-TDS modeling for the first time. Through a comprehensive analysis of forward problems in the time domain, we demonstrate the feasibility and potential of PINNs as a powerful framework for advancing THz wave simulation and analysis.

physics.optics↗

arXiv2025

Modeling Terahertz Propagation via Frequency-Domain Physics-Informed Neural Networks

Terahertz time-domain spectroscopy (THz-TDS) provides a non-invasive and label-free method for probing the internal structure and electromagnetic response of materials. Numerical simulation of THz-TDS can help understanding wave-matter interactions, guiding experimental design, and interpreting complex measurement data. However, existing simulation techniques face challenges in accurately modeling THz wave propagation with low computational cost. Additionally, conventional simulation solvers often require dense spatial-temporal discretization, which limits their applicability to large-scale and real-time scenarios. Simplified analytical models may neglect dispersion, multiple scattering, and boundary effects. To address these limitations, we establish a novel computational framework that integrates frequency-domain physics-informed neural networks (FD-PINNs) with less data-driven. To validate our proposed FD-PINNs, simulation results from finite-difference time-domain (FDTD) and time-domain (TD)-PINNs were used to compare with FD-PINNs. Finally, experimental results from THz-TDS systems were employed to further exhibit accurate reconstruction ability of FD-PINNs.

physics.optics↗

arXiv2025

Thermal diffusivity measurement based on evaporative cryocooling excitation: Theory and experiments

Photo-thermal methods for measuring thermal diffusivity inherently pose an ill-posed inverse problem, affected by factors such as sample thickness, heating or cooling time, and excitation energy. Measurement accuracy becomes particularly challenging under non-impulsive pulsed excitation when the observation timescale is comparable to the pulse duration. This is often due to poorly defined pulse shapes, broadened thermal responses, and the absence of clear boundary conditions, especially under significant interfacial temperature gradients where natural convection dominates. The classic Parker solution, while widely used, is physically unrealistic as it assumes adiabatic heat flux and shallow-region heat absorption. In this study, we prove that Parker's assumption is equivalent to the Dirac pulse boundary condition in mathematics. Then, we present comprehensive analytical solutions for thermal/cooling responses under Dirac and rectangular pulse excitations. By comparing with eigenfunction-based solutions for well-posed boundary conditions, we show that Parker's solution is only valid before the thermal peak. Through dimensionless processing, we further demonstrate that Parker's solution can be regarded as a limiting case of the rectangular pulse solution as the heating duration approaches zero. Furthermore, we propose a novel excitation approach, evaporative cryocooling, for thermal diffusivity measurement. This method offers a compact, low-cost, and easy-to-implement alternative to conventional excitation schemes. The theoretical model was further validated through comparison with experimental results.

physics.app-ph↗

arXiv2025

A novel IR-SRGAN assisted super-resolution evaluation of photothermal coherence tomography for impact damage in toughened thermoplastic CFRP laminates under room temperature and low temperature

Evaluating impact-induced damage in composite materials under varying temperature conditions is essential for ensuring structural integrity and reliable performance in aerospace, polar, and other extreme-environment applications. As matrix brittleness increases at low temperatures, damage mechanisms shift: impact events that produce only minor delaminations at ambient conditions can trigger extensive matrix cracking, fiber/matrix debonding, or interfacial failure under severe cold loads, thereby degrading residual strength and fatigue life. Precision detection and quantification of subsurface damage features (e.g., delamination area, crack morphology, interface separation) are critical for subsequent mechanical characterization and life prediction. In this study, infrared thermography (IRT) coupled with a newly developed frequency multiplexed photothermal correlation tomography (FM-PCT) is employed to capture three-dimensional subsurface damage signatures with depth resolution approaching that of X-ray micro-computed tomography. However, the inherent limitations of IRT, including restricted frame rate and lateral thermal diffusion, reduce spatial resolution and thus the accuracy of damage size measurement. To address this, we develop a new transfer learning-based infrared super-resolution generative adversarial network (IR-SRGAN) that enhances both lateral and depth-resolved imaging fidelity based on limited thermographic datasets.

physics.app-ph↗

arXiv2025

Thermal diffusivity characterization of impacted composites using evaporative cryocooling excitation and inverse physics-informed neural networks

The thermal diffusivity measurement of impacted composites using pulsed methods presents an ill-posed inverse problem influenced by multiple factors such as sample thickness, cooling duration, and excitation energy. In this study, a novel excitation method, evaporative cryocooling, was introduced for measuring the thermal diffusivity of tested samples. Compared to conventional excitation modalities, evaporative cryocooling excitation is compact, portable, and low cost. However, evaporative cryocooling cannot be considered a pulsed method due to its prolonged excitation duration. In general, it is difficult to measure thermal diffusivity based on non-impulsive pulsed excitation at times commensurate with the pulse duration, often due to ill-defined pulse shape and width and the subsequent potentially complicated thermal response which may be subject to diffusive broadening. To address this challenge, inverse physics-informed neural networks (IPINNs) were introduced in this work and integrated with an evaporative cryocooling method. The Parker method combined with a photothermal method was employed as a reference. To improve the accuracy of both IPINNs and Parker methods, terahertz time-domain spectroscopy (THz-TDS) was employed for measuring the thickness of impacted composites. Simulations and experimental results demonstrated the feasibility and accuracy of the IPINN-based approach.

physics.app-ph↗