检索结果-内蒙古大学图书馆

HSI-MFormer: Integrating Mamba and Transformer Experts for Hyperspectral Image Classification

学校读者我要写书评

暂无评论

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： He, Yan Tu, Bing Liu, Bo Li, Jun Plaza, Antonio Nanjing University of Information Science and Technology Institute of Optics and Electronics State Key Laboratory Cultivation Base of Atmospheric Optoelectronic Detection and Information Fusion Jiangsu International Joint Laboratory on Meteorological Photonics and Optoelectronic Detection Jiangsu Engineering Research Center for Intelligent Optoelectronic Sensing Technology of Atmosphere Nanjing210044 China China University of Geosciences Faculty of Computer Science Wuhan430074 China University of Extremadura Hyperspectral Computing Laboratory Department of Technology of Computers and Communications Escuela Politecnica Cáceres‎10003 Spain

Hyperspectral image (HSI) classification is fundamental to numerous remote sensing applications, enabling detailed analysis of material properties and environmental conditions. Recent Mamba built upon selective state space models (S6) have demonstrated exceptional advantages in long-range sequence modeling with linear computational efficiency, while Transformer based on self-attention mechanisms is particularly adept at capturing short-range dependencies. To leverage the complementary strengths of these models, this paper introduces a novel hybrid Mamba-Transformer framework (HSI-MFormer), effectively exploring the multiscale properties of hyperspectral data for HSI classification. Initially, a Multiscale Token Generation module (MTG) is developed, which converts the HSI cube into multiple spatial-spectral token groups across different scales. To adequately capture fine-grained multiscale spatial-spectral patterns, an Inner-scale Transformer Expert (ITE) is designed, which incorporates grouped self-attention operations to perform short-range sequence modeling within token groups at each scale. Meanwhile, a Cross-scale Mamba Expert (CME) is introduced, which integrates a cross-scale serialization mechanism and bidirectional Mamba block for long-range sequence modeling, further exploring the interactions and complementarity between token groups across different scales. Several hybrid strategies for integrating the ITE and CME are investigated to maximize their complementarity, including parallel, interval, and serial structures. Extensive experiments demonstrate that the propsed HSI-MFormer significantly out-performs the state-of-the-art Transformer-based and Mamba-based HSI classification methods. © 1980-2012 IEEE.

关键词： Image classification

CLIMS++: Cross Language Image Matching with Automatic Context Discovery for Weakly Supervised Semantic Segmentation

学校读者我要写书评

暂无评论

International Journal of Computer Vision 2025年 1-20页

作者： Xie, Jinheng Deng, Songhe Hou, Xianxu Luo, Zhaochuan Shen, Linlin Huang, Yawen Zheng, Yefeng Shou, Mike Zheng Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Show Lab National University of Singapore Singapore Singapore School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou China Department of Computer Science Wenzhou-Kean University Wenzhou China Jarvis Research Center Tencent YouTu Lab Shenzhen China

While promising results have been achieved in weakly-supervised semantic segmentation (WSSS), limited supervision from image-level tags inevitably induces discriminative reliance and spurious relations between target classes and background regions. Thus, Class Activation Map (CAM) usually tends to activate discriminative object regions and falsely includes lots of class-related backgrounds. Without pixel-level supervisions, it could be very difficult to enlarge the foreground activation and suppress those false activation of background regions. In this paper, we propose a novel framework of Cross Language Image Matching with Automatic Context Discovery (CLIMS++), based on the recently introduced Contrastive Language-Image Pre-training (CLIP) model, for WSSS. The core idea of our framework is to introduce natural language supervision to activate more complete object regions and suppress class-related background regions in CAM. In particular, we design object, background region, and text label matching losses to guide the model to excite more reasonable object regions of each category. In addition, we propose to automatically find spurious relations between foreground categories and backgrounds, through which a background suppression loss is designed to suppress the activation of class-related backgrounds. The above designs enable the proposed CLIMS++ to generate a more complete and compact activation map for the target objects. Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 datasets show that our CLIMS++ significantly outperforms the previous state-of-the-art methods.

关键词： Multi-modal learning Semantic segmentation Weakly-supervised learning

Enhanced Large Language Models for Effective Screening of Depression and Anxiety

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liu, June M. Gao, Mengxia Sabour, Sahand Chen, Zhuang Huang, Minlie Lee, Tatia M.C. State Key Laboratory of Brain and Cognitive Sciences The University of Hong Kong Hong Kong Laboratory of Neuropsychology and Human Neuroscience The University of Hong Kong Hong Kong The CoAI group DCST Institute for Artificial Intelligence State Key Lab of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology Tsinghua University Beijing China

Depressive and anxiety disorders are widespread, necessitating timely identification and management. Recent advances in Large Language Models (LLMs) offer potential solutions, yet high costs and ethical concerns about training data remain challenges. This paper introduces a pipeline for synthesizing clinical interviews, resulting in 1,157 interactive dialogues (PsyInterview), and presents EmoScan, an LLM-based emotional disorder screening system. EmoScan distinguishes between coarse (e.g., anxiety or depressive disorders) and fine disorders (e.g., major depressive disorders) and conducts high-quality interviews. Evaluations showed that EmoScan exceeded the performance of base models and other LLMs like GPT-4 in screening emotional disorders (F1-score=0.7467). It also delivers superior explanations (BERTScore=0.9408) and demonstrates robust generalizability (F1-score of 0.67 on an external dataset). Furthermore, EmoScan outperforms baselines in interviewing skills, as validated by automated ratings and human evaluations. This work highlights the importance of scalable data-generative pipelines for developing effective mental health LLM tools. © 2025, CC BY-NC-SA.

关键词：

AiCareGaitRehabilitation: Multi-modalities sensor data fusion for AI-IoT enabled realtime electrical stimulation device for pre-Fog and post-Fog to person with Parkinson's Disease

学校读者我要写书评

暂无评论

Information Fusion 2025年 122卷

作者： Ghayvat, Hemant Awais, Muhammad Geddam, Rebakah Quasim, Mohammad Tabrez Khowaja, Sunder Ali Dev, Kapal Computer Science and Media Technology Department Faculty of Technology Linnaeus University P G Vejdes väg 351 95 Växjö35234 Sweden Departments of Medical Physics Memorial Sloan Kettering Cancer Center New YorkNY10065 United States Unitedworld Institute of Technology Karnavati University Gujarat Gandhinagar382422 India Department of Computer Science and Artificial Intelligence College of Computing and Information Technology University of Bisha Bisha551 Saudi Arabia ADAPT Centre and School of Computing Dublin City University DublinD09 V209 Ireland ADAPT Centre and Department of Computer Science Munster Technological University Bishopstown CorkT12 P928 Ireland Department of Institute of intelligent Systems University of Johannesburg Auckland Park JohannesburgT12 P928 South Africa Center for Research Impact & Outcome Chitkara University Institute of Engineering and Technology Rapura Punjab 140401 India

The neuromuscular dysfunction known as Freezing of Gait (FoG), which is more prevalent in individuals suffering from Parkinson's Disease (PD), significantly reduces the quality of life and increases their risk of falling. Wearable FoG sensing technologies provide timely biofeedback cues to assist people regain control over their gait. However, the devices being bulky, intrusive, and annoyance of current FoG detection algorithms limit their usability in real-world applications. This study proposes a more efficient approach by integrating the FoG detection fusion algorithm into a Functional Electrical Stimulation (FES) module. The design leverages features with low computational requirements and specialized hardware to minimize the use of physical space and memory. The Convolutional Neural Networks (CNN) approach with SVM output was deployed to classify FoG and non-FoG periods in real-time. Additionally, the study uses CNN algorithms in fusion with data from a triaxial accelerometer, strain sensors, and piezoelectric plantar sensors to test shank-worn FoG detection devices. The study demonstrates that electrical stimulation-based cueing strategies significantly improve gait control and mitigate FoG episodes in people with Parkinson's disease. The AiCareGaitRehabilitation system employs a multi-modal sensor fusion strategy to improve the efficacy of the FES device. Data from various sensors—such as strain sensor, 18 plantar sensors, and four quadriceps sensors—the system provides a holistic view of both pre-freezing of Gait (pre-FOG) and post-freezing of Gait (post-FOG) scenarios. This research aims to improve mobility, reduce fall risks, and eventually improve the quality of life for individuals with Parkinson's disease. © 2025

关键词： Gait analysis

Balanced Multi-view Clustering

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Li, Zhenglai Wang, Jun Tang, Chang Zhu, Xinzhong Zhang, Wei Liu, Xinwang Faculty of data science City University of Macau 999078 China School of computer National University of Defense Technology Changsha410073 China School of computer China University of Geosciences Wuhan430074 China College of Mathematics Physics and Information Engineering Zhejiang Normal University Jinhua China Research Institute of Ningbo Cixing Co. Ltd Ningbo China Qilu University of Technology Shandong Academy of Sciences China Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Shandong Fundamental Research Center for Computer Science Jinan250000 China

Multi-view clustering (MvC) aims to integrate information from different views to enhance the capability of the model in capturing the underlying data structures. The widely used joint training paradigm in MvC is potentially not fully leverage the multi-view information, since the imbalanced and under-optimized view-specific features caused by the uniform learning objective for all views. For instance, particular views with more discriminative information could dominate the learning process in the joint training paradigm, leading to other views being under-optimized. To alleviate this issue, we first analyze the imbalanced phenomenon in the joint-training paradigm of multi-view clustering from the perspective of gradient descent for each view-specific feature extractor. Then, we propose a novel balanced multi-view clustering (BMvC) method, which introduces a view-specific contrastive regularization (VCR) to modulate the optimization of each view. Concretely, VCR preserves the sample similarities captured from the joint features and view-specific ones into the clustering distributions corresponding to view-specific features to enhance the learning process of view-specific feature extractors. Additionally, a theoretical analysis is provided to illustrate that VCR adaptively modulates the magnitudes of gradients for updating the parameters of view-specific feature extractors to achieve a balanced multi-view learning procedure. In such a manner, BMvC achieves a better trade-off between the exploitation of view-specific patterns and the exploration of view-invariance patterns to fully learn the multi-view information for the clustering task. Finally, a set of experiments are conducted to verify the superiority of the proposed method compared with state-of-the-art approaches on eight benchmark MvC datasets. The demo code of this work is publicly available at https://***/guanyuezhen/BMvC. Copyright © 2025, The Authors. All rights reserved.

关键词： Contrastive Learning

Environment-Driven Online LiDAR-Camera Extrinsic Calibration

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Huang, Zhiwei Li, Jiaqi Zhong, Ping Fan, Rui The Department of Control Science & Engineering The College of Electronics & Information Engineering Tongji University Shanghai201804 China The School of Computer Science and Engineering Central South University Hunan Changsha410083 China The National Key Laboratory of Science and Technology on Automatic Target Recognition National University of Defense Technology Hunan Changsha410073 China The Department of Control Science & Engineering The College of Electronics & Information Engineering Shanghai Research Institute for Intelligent Autonomous Systems The State Key Laboratory of Intelligent Autonomous Systems Frontiers Science Center for Intelligent Autonomous Systems Tongji University Shanghai201804 China The National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Shaanxi Xi’an710049 China

LiDAR-camera extrinsic calibration (LCEC) is the core for data fusion in computer vision. Existing methods typically rely on customized calibration targets or fixed scene types, lacking the flexibility to handle variations in sensor data and environmental contexts. This paper introduces EdO-LCEC, the first environment-driven, online calibration approach that achieves human-like adaptability. Inspired by the human perceptual system, EdO-LCEC incorporates a generalizable scene discriminator to actively interpret environmental conditions, creating multiple virtual cameras that capture detailed spatial and textural information. To overcome cross-modal feature matching challenges between LiDAR and camera, we propose dual-path correspondence matching (DPCM), which leverages both structural and textural consistency to achieve reliable 3D-2D correspondences. Our approach formulates the calibration process as a spatial-temporal joint optimization problem, utilizing global constraints from multiple views and scenes to improve accuracy, particularly in sparse or partially overlapping sensor views. Extensive experiments on real-world datasets demonstrate that EdO-LCEC achieves state-of-the-art performance, providing reliable and precise calibration across diverse, challenging environments. Copyright © 2025, The Authors. All rights reserved.

关键词： Spatio-temporal data

DMSACNN: Deep Multiscale Attentional Convolutional Neural Network for EEG-Based Motor Decoding

学校读者我要写书评

暂无评论

IEEE Journal of Biomedical and Health Informatics 2025年 PP卷 PP页

作者： Liu, Ke Xing, Xin Yang, Tao Yu, Zhuliang Xiao, Bin Wang, Guoyin Wu, Wei Chongqing University of Posts and Telecommunications School of Computer Science and Technology Chongqing400065 China Chongqing University of Posts and Telecommunications Key Laboratory of Big Data Intelligent Computing China Chongqing University of Posts and Telecommunications Key Laboratory of Cyberspace Big Data Intelligent Security of Ministry of Education China South China University of Technology College of Automation Science and Engineering Guangzhou510641 China Chongqing Normal University National Center for Applied Mathematics in Chongqing Chongqing401331 China Shanghai Jiao Tong University School of Medicine Songjiang Hospital Songjiang Research Institute Shanghai Key Laboratory of Emotions and Affective Disorders Shanghai201600 China

Objective: Accurate decoding of electroencephalogram (EEG) signals has become more significant for the brain-computer interface (BCI). Specifically, motor imagery and motor execution (MI/ME) tasks enable the control of external devices by decoding EEG signals during imagined or real movements. However, accurately decoding MI/ME signals remains a challenge due to the limited utilization of temporal information and ineffective feature selection methods. Methods: This paper introduces DMSACNN, an end-to-end deep multiscale attention convolutional neural network for MI/ME-EEG decoding. DMSACNN incorporates a deep multiscale temporal feature extraction module to capture temporal features at various levels. These features are then processed by a spatial convolutional module to extract spatial features. Finally, a local and global feature fusion attention module is utilized to combine local and global information and extract the most discriminative spatiotemporal features. Main results: DMSACNN achieves impressive accuracies of 78.20%, 96.34% and 70.90% for hold-out analysis on the BCI-IV-2a, High Gamma and OpenBMI datasets, respectively, outperforming most of the state-of-the-art methods. Conclusion and significance: These results highlight the potential of DMSACNN in robust BCI applications. Our proposed method provides a valuable solution to improve the accuracy of the MI/ME-EEG decoding, which can pave the way for more efficient and reliable BCI systems. The source code for DMSACNN is available at https://***/xingxin-99/***. © 2025 IEEE.

关键词： Convolutional neural networks

SFedXL: Semi-Synchronous Federated Learning With Cross-Sharpness and Layer-Freezing

学校读者我要写书评

暂无评论

IEEE Internet of Things Journal 2025年

作者： Zhao, Mingxiong Zhao, Shihao Feng, Chenyuan Yang, Howard H. Niyato, Dusit Quek, Tony Q. S. Yunnan University National Pilot School of Software Kunming650500 China Ministry of Education Engineering Research Center of Integration and Application of Digital Learning Technology Beijing100039 China Ministry of Education Engineering Research Center of Cyberspace Kunming650504 China Yunnan University of Finance and Economics Yunnan Key Laboratory of Service Computing Kunming650221 China EURECOM Sophia Antipolis06410 France Zhejiang University University of Illinois Urbana-Champaign Institute Zhejiang University Haining314400 China Nanyang Technological University School of Computer Science and Engineering 639798 Singapore Singapore University of Technology and Design Information Systems Technology and Design Pillar 487372 Singapore

Federated learning (FL) emerges as a potential solution for enabling multiple terminal devices to collaboratively accomplish computational tasks within an Unmanned Aerial Vehicle (UAV) swarm. However, traditional FL approaches, predicated on synchronous data aggregation, are not feasible for a UAV swarm owing to the inherently variable and dynamic nature of their communication networks compared with terrestrial systems. Furthermore, the data procured by UAVs is often highly heterogeneous, attributable to disparities in deployment environments and device attributes. Considering the distinct flight paths and unique operational conditions encountered by different UAVs, a considerable amount of data remains unlabeled. To tackle the challenges associated with asynchronous operations and the prevalence of unlabeled data, we introduce a novel framework termed Semi-synchronous FL with Cross-Sharpness and Layer-Freezing (SFedXL), tailored for a UAV swarm. In particular, we devise a cross-sharpness model training strategy aimed at optimizing the utilization of both labeled and unlabeled datasets. Additionally, we propose an innovative semi-synchronous model aggregation protocol, complemented by client-specific layer-freezing and client cluster scheduling, designed to expedite the training process. Our simulation results indicate that the proposed algorithm surpasses current FL methods in terms of object recognition accuracy and communication efficiency, albeit with a trade-off of increased local computation latency. © 2014 IEEE.

关键词： Federated learning

Rapid quantum ground state preparation via dissipative dynamics

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhan, Yongtao Ding, Zhiyan Huhn, Jakob Gray, Johnnie Preskill, John Chan, Garnet Kin-Lic Lin, Lin Institute for Quantum Information and Matter California Institute of Technology United States Division of Physics Mathematics and Astronomy California Institute of Technology United States Department of Mathematics University of California Berkeley United States Department of Physics Arnold Sommerfeld Center for Theoretical Physics Ludwig-Maximilians-Universität München Germany Division of Chemistry and Chemical Engineering California Institute of Technology United States Department of Physics California Institute of Technology United States AWS Center for Quantum Computing United States Applied Mathematics and Computational Research Division Lawrence Berkeley National Laboratory United States

Inspired by natural cooling processes, dissipation has become a promising approach for preparing low-energy states of quantum systems. However, the potential of dissipative protocols remains unclear beyond certain commuting Hamiltonians. This work provides significant analytical and numerical insights into the power of dissipation for preparing the ground state of non-commuting Hamiltonians. For quasi-free dissipative dynamics, including certain 1D spin systems with boundary dissipation, our results reveal a new connection between the mixing time in trace distance and the spectral properties of a non-Hermitian Hamiltonian, leading to an explicit and sharp bound on the mixing time that scales polynomially with system size. For more general spin systems, we develop a tensor network-based algorithm for constructing the Lindblad jump operator and for simulating the dynamics. Using this algorithm, we demonstrate numerically that dissipative ground state preparation protocols can achieve rapid mixing for certain 1D local Hamiltonians under bulk dissipation, with a mixing time that scales logarithmically with the system size. We then prove the rapid mixing result for certain weakly interacting spin and fermionic systems in arbitrary dimensions, extending recent results for high-temperature quantum Gibbs samplers to the zero-temperature regime. Our theoretical approaches are applicable to systems with singular stationary states, and are thus expected to have applications beyond the specific systems considered in this study. © 2025, CC BY.

关键词： Hamiltonians