检索结果-内蒙古大学图书馆

Spiking Neural Networks for Temporal Processing: Status Quo and Future Prospects

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ma, Chenxiang Chen, Xinyi Li, Yanchen Yang, Qu Wu, Yujie Li, Guoqi Pan, Gang Tang, Huajin Tan, Kay Chen Wu, Jibin Department of Data Science and Artificial Intelligence The Hong Kong Polytechnic University Hong Kong Department of Computing The Hong Kong Polytechnic University Hong Kong Department of Electrical and Computer Engineering National University of Singapore 119077 Singapore Institute of Automation Chinese Academy of Sciences Beijing100045 China State Key Laboratory of Brain-Machine Intelligence College of Computer Science and Technology MOE Frontier Science Center for Brain Science and Brain-Machine Integration Zhejiang University Hangzhou310027 China

Temporal processing is fundamental for both biological and artificial intelligence systems, as it enables the comprehension of dynamic environments and facilitates timely responses. Spiking Neural Networks (SNNs) excel in handling such data with high efficiency, owing to their rich neuronal dynamics and sparse activity patterns. Given the recent surge in the development of SNNs, there is an urgent need for a comprehensive evaluation of their temporal processing capabilities. In this paper, we first conduct an in-depth assessment of commonly used neuromorphic benchmarks, revealing critical limitations in their ability to evaluate the temporal processing capabilities of SNNs. To bridge this gap, we further introduce a benchmark suite consisting of three temporal processing tasks characterized by rich temporal dynamics across multiple timescales. Utilizing this benchmark suite, we perform a thorough evaluation of recently introduced SNN approaches to elucidate the current status of SNNs in temporal processing. Our findings indicate significant advancements in recently developed spiking neuron models and neural architectures regarding their temporal processing capabilities, while also highlighting a performance gap in handling long-range dependencies when compared to state-of-the-art non-spiking models. Finally, we discuss the key challenges and outline potential avenues for future research. Copyright © 2025, The Authors. All rights reserved.

关键词： Spatio-temporal data

DCSENets: Interpretable deep learning for patient-independent seizure classification using enhanced EEG-based spectrogram visualization

学校读者我要写书评

暂无评论

Computers in Biology and Medicine 2025年 185卷 109558-109558页

作者： Aboyeji, Sunday Timothy Ahmad, Ijaz Wang, Xin Chen, Yan Yao, Chen Li, Guanglin Tong, Michael Chi Fai Siu, Alice K.Y. Zhao, Guoru Chen, Shixiong CAS Key Laboratory of Human–Machine Intelligence Synergy Systems Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Shenzhen518055 China University of Chinese Academy of Sciences Beijing100049 China Guangdong-Hong Kong-Macao Joint Laboratory of Human–Machine Intelligence-Synergy Systems Guangdong Shenzhen518055 China Surgery Division Epilepsy Center Shenzhen Children's Hospital Shenzhen518036 China Department of Neurosurgery The National Key Clinic Specialty Shenzhen Key Laboratory of Neurosurgery the First Affiliated Hospital of Shenzhen University Shenzhen Second People's Hospital Shenzhen518035 China Department of Otorhinolaryngology Head and Neck Surgery The Chinese University of Hong Kong Hong Kong School of Medicine The Chinese University of Hong Kong Shenzhen Guangdong518172 China

Neurologists often face challenges in identifying epileptic activities within multichannel EEG recordings, requiring extensive hours of analysis. Computer-aided diagnosis systems have been proposed to reduce manual inspection of EEG signals by neurologists. However, direct analysis of EEG signals is difficult due to their complex and dynamic nature, with variation across multiple patients. Therefore, researchers have proposed the short-time Fourier transform (STFT) to capture dynamic events indicative of seizures through time-varying frequency representation of EEG signals. However, tradeoffs between time and frequency resolution limited the spectrogram's interpretability and affected clinical deployment. Hence, this study proposes extracting high-resolution channels via a novel STFT spectrogram construction algorithm encompassing taper functions for seizure diagnosis. Initially, we extracted seizure and non-seizure segments from each channel of selected patients in the CHB-MIT dataset. Next, we systematically apply taper functions like Hann and Gaussian windows to minimize the edge effect during the construction of spectrogram images. Finally, we employ Dilated Convolutional Squeeze and Excitation Networks (DCSENets) through leave-one-patient-out cross-validation (LOPOCV) to perform patient-independent seizure classification. The proposed DCSENets achieve an average accuracy of 87.20±11.48% and 87.29±10.48% with Hann and Gaussian taper functions, respectively, and 86.85±11.56% without the taper function. Most patients with high performances indicate similarity in train-test sample distribution using the Kolmogorov–Smirnov test at 0.010.05. Furthermore, the Grad CAM deep visual explainer integration enhances the interpretability of the deep learning model's decision-making process. Consequently, neurologists are provided not only with enhanced visualized spectrograms but also a transparent model for improved seizure diagnosis. © 2024 Elsevier Ltd

关键词： Spectrographs

High Efficiency Wiener Filter-based Point Cloud Quality Enhancement for MPEG G-PCC

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wei, Yuxuan Wang, Zehan Guo, Tian Liu, Hao Shen, Liquan Yuan, Hui School of Control Science and Engineering Shandong University Jinan250061 China Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan250061 China School of Computer and Control Engineering Yantai University Yantai264005 China Shanghai Institute for Advanced Communication and Data Science Shanghai University Shanghai200072 China

Point clouds, which directly record the geometry and attributes of scenes or objects by a large number of points, are widely used in various applications such as virtual reality and immersive communication. However, due to the huge data volume and unstructured geometry, efficient compression of point clouds is very crucial. The Moving Picture Expert Group is establishing a geometry-based point cloud compression (G-PCC) standard for both static and dynamic point clouds in recent years. Although lossy compression of G-PCC can achieve a very high compression ratio, the reconstruction quality is relatively low, especially at low bitrates. To mitigate this problem, we propose a high efficiency Wiener filter that can be integrated into the encoder and decoder pipeline of G-PCC to improve the reconstruction quality as well as the rate-distortion performance for dynamic point clouds. Specifically, we first propose a basic Wiener filter, and then improve it by introducing coefficients inheritance and variance-based point classification for the Luma component. Besides, to reduce the complexity of the nearest neighbor search during the application of the Wiener filter, we also propose a Morton code-based fast nearest neighbor search algorithm for efficient calculation of filter coefficients. Experimental results demonstrate that the proposed method can achieve average Bjøntegaard delta rates of -6.1%, -7.3%, and -8.0% for Luma, Chroma Cb, and Chroma Cr components, respectively, under the condition of lossless-geometry-lossy-attributes configuration compared to the latest G-PCC encoding platform (i.e., geometry-based solid content test model version 7.0 release candidate 2) by consuming affordable computational complexity. © 2025, CC BY.

关键词： Wiener filtering

STMDNet: A Lightweight Directional Framework for Motion Pattern Recognition of Tiny Targets

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Xu, Mingshuo Luan, Hao Hao, Zhou Daniel Peng, Jigen Yue, Shigang The School of Mathematics and Computing Science University of Leicester LeicesterLE1 7RH United Kingdom The Tianjin Key Laboratory of Information Sensing and Intelligent Control School of Automation and Electrical Engineering Tianjin University of Technology and Education Tuanjin 300222 China The Machine Life and Intelligence Research Center Guangzhou University Guangzhou510006 China

Recognizing motions of tiny targets—only few dozen pixels—in cluttered backgrounds remains a fundamental challenge when standard feature-based or deep learning methods fail under scarce visual cues. We propose STMDNet, a model-based computational framework to Recognize motions of tiny targets at variable velocities under low-sampling frequency scenarios. STMDNet designs a novel dual-dynamics-and-correlation mechanism, harnessing ipsilateral excitation to integrate target cues and leakage-enhancing-type contralateral inhibition to suppress large-object and background motion interference. Moreover, we develop the first collaborative directional encoding-decoding strategy that determines the motion direction from only one correlation per spatial location, cutting computational costs to one-eighth of prior methods. Further, simply substituting the backbone of a strong STMD model with STMDNet raises AUC by 24%, yielding an enhanced STMDNet-F. Evaluations on real-world low-sampling frequency datasets show state-of-the-art results, surpassing the deep learning baseline. Across diverse speeds, STMDNet-F improves mF1 by 19%, 16%, and 8% at 240Hz, 120Hz, and 60Hz, respectively, while STMDNet achieves 87 FPS on a single CPU thread. These advances highlight STMDNet as a next-generation backbone for tiny target motion pattern recognition and underscore its broader potential to revitalize model-based visual approaches in motion detection. Code is available at https://***/MingshuoXu/STMDNet. Copyright © 2025, The Authors. All rights reserved.

关键词： Signal encoding

Shape Embedding and Knowledge Mining Network for Generalized Few-Shot Remote Sensing Segmentation

学校读者我要写书评

暂无评论

IEEE Geoscience and Remote Sensing Letters 2025年

作者： Qiu, Zifeng Liu, Hongyu Xiong, Hang Di, Chengliang Fang, Hao Cong, Runmin Beijing Jiaotong University Institute of Information Science Beijing100044 China The54th Research Institute of CETC Shijiazhuang050081 China China Communication System Co. Ltd. Hebei Branch Shijiazhuang050081 China Shandong University School of Control Science and Engineering Jinan250061 China Ministry of Education Key Laboratory of Machine Intelligence and System Control Jinan250061 China Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing100044 China

In recent years, generalized few-shot segmentation (GFSS) has received widespread attention from scholars by virtue of its superiority in low-data regimes. Most of the existing research focuses on natural image processing, and few studies have been devoted to the practical but challenging topic of remote sensing image (RSI) understanding. In this paper, we propose a Shape Embedding and Knowledge Mining Network (SKNet) for generalized few-shot RSI segmentation. Specifically, the framework is divided into two key stages: (a) In the base class learning stage, shape representation embedding is introduced to enhance the network's ability to perceive remote sensing objects. Simultaneously, we introduce the self-reconstruction constraint to prevent new unseen classes from merging, thereby improving the representation uniqueness of these classes. (b) In the novel class learning stage, a base class knowledge mining mechanism is designed to update the prototypes of the novel class by using the prototype representation of the base class, so as to enhance the discrimination ability of the network. We validated our methods on the adapted version of OpenEarthMap and iSAID datasets. In comparison with existing GFSS methods, the proposed approach demonstrates an advancement. © 2004-2012 IEEE.

关键词： Semantic Segmentation

CAD-Unet: A capsule network-enhanced Unet architecture for accurate segmentation of COVID-19 lung infections from CT images

学校读者我要写书评

暂无评论

Medical Image Analysis 2025年 103卷 103583页

作者： Dang, Yijie Ma, Weijun Luo, Xiaohu Wang, Huaizhu School of Information Engineering Ningxia University Yinchuan Ningxia750021 China Ningxia Key Laboratory of Artificial Intelligence and Information Security for Channeling Computing Resources from the East to the West Ningxia University Yinchuan Ningxia750021 China School of Mathematics and Computer Science Ningxia Normal University Guyuan756099 China School of Advanced Interdisciplinary Studies Ningxia University Zhongwei755000 China

Since the outbreak of the COVID-19 pandemic in 2019, medical imaging has emerged as a primary modality for diagnosing COVID-19 pneumonia. In clinical settings, the segmentation of lung infections from computed tomography images enables rapid and accurate quantification and diagnosis of COVID-19. Segmentation of COVID-19 infections in the lungs poses a formidable challenge, primarily due to the indistinct boundaries and limited contrast presented by ground glass opacity manifestations. Moreover, the confounding similarity among infiltrates, lung tissues, and lung walls further complicates this segmentation task. To address these challenges, this paper introduces a novel deep network architecture, called CAD-Unet, for segmenting COVID-19 lung infections. In this architecture, capsule networks are incorporated into the existing Unet framework. Capsule networks represent a novel type of network architecture that differs from traditional convolutional neural networks. They utilize vectors for information transfer among capsules, facilitating the extraction of intricate lesion spatial information. Additionally, we design a capsule encoder path and establish a coupling path between the unet encoder and the capsule encoder. This design maximizes the complementary advantages of both network structures while achieving efficient information fusion. Finally, extensive experiments are conducted on four publicly available datasets, encompassing binary segmentation tasks and multi-class segmentation tasks. The experimental results demonstrate the superior segmentation performance of the proposed model. The code has been released at: https://***/AmanoTooko-jie/CAD-Unet. © 2025 Elsevier B.V.

关键词： COVID-19

Decoupled Distillation to Erase: A General Unlearning Method for Any Class-centric Tasks

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhou, Yu Zheng, Dian Mo, Qijie Lu, Renjie Lin, Kun-Yu Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Peng Cheng Laboratory China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Guangdong Province Key Laboratory of Information Security Technology China

In this work, we present DEcoupLEd Distillation To Erase (DELETE), a general and strong unlearning method for any class-centric tasks. To derive this, we first propose a theoretical framework to analyze the general form of unlearning loss and decompose it into forgetting and retention terms. Through the theoretical framework, we point out that a class of previous methods could be mainly formulated as a loss that implicitly optimizes the forgetting term while lacking supervision for the retention term, disturbing the distribution of pre-trained model and struggling to adequately preserve knowledge of the remaining classes. To address it, we refine the retention term using "dark knowledge" and propose a mask distillation unlearning method. By applying a mask to separate forgetting logits from retention logits, our approach optimizes both the forgetting and refined retention components simultaneously, retaining knowledge of the remaining classes while ensuring thorough forgetting of the target class. Without access to the remaining data or intervention (i.e., used in some works), we achieve state-of-the-art performance across various benchmarks. What’s more, DELETE is a general solution that can be applied to various downstream tasks, including face recognition, backdoor defense, and semantic segmentation with great performance. Copyright © 2025, The Authors. All rights reserved.

关键词：

Modeling Multiple Normal Action Representations for Error Detection in Procedural Tasks

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Huang, Wei-Jin Li, Yuan-Ming Xia, Zhi-Wei Tang, Yu-Ming Lin, Kun-Yu Hu, Jian-Fang Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Peng Cheng Laboratory China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China Guangdong Province Key Laboratory of Information Security Technology China

Error detection in procedural activities is essential for consistent and correct outcomes in AR-assisted and robotic systems. Existing methods often focus on temporal ordering errors or rely on static prototypes to represent normal actions. However, these approaches typically overlook the common scenario where multiple, distinct actions are valid following a given sequence of executed actions. This leads to two issues: (1) the model cannot effectively detect errors using static prototypes when the inference environment or action execution distribution differs from training;and (2) the model may also use the wrong prototypes to detect errors if the ongoing action label is not the same as the predicted one. To address this problem, we propose an Adaptive Multiple Normal Action Representation (AMNAR) framework. AMNAR predicts all valid next actions and reconstructs their corresponding normal action representations, which are compared against the ongoing action to detect errors. Extensive experiments demonstrate that AMNAR achieves state-of-the-art performance, highlighting the effectiveness of AMNAR and the importance of modeling multiple valid next actions in error detection. The code is available at https://***/iSEElaboratory/AMNAR. Copyright © 2025, The Authors. All rights reserved.

关键词： Error detection

PolarBEVU: Multi-Camera 3D Object Detection in Polar Bird's-Eye View via Unprojection

学校读者我要写书评

暂无评论

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Hou, Minghui Lyu, Chuanhao Wang, Gang Ma, Baorui Xu, Rongtao Hu, Jue Fan, Xiaopeng Jilin University College of Computer Science and Technology Changchun China Tsinghua University School of Software Beijing China Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Beijing China Harbin Institute of Technology National Key Laboratory of Science and Technology on Advanced Composites in Special Environments Harbin150080 China Harbin Institute of Technology Centre for Composite Materials and Structures Harbin150001 China Harbin Institute of Technology Faculty of Computing Harbin150001 China

3D object detection from a Bird's Eye View (BEV) has emerged as a novel perception paradigm for autonomous driving scenarios. While most current 3D object detection methods still rely on the conventional Cartesian coordinates, they fail to align with the non-aligned coordinate system inherent in image geometry. The Polar coordinates, on the other hand, better fit with the geometric shape corresponding to the perception of cameras. However, transforming between coordinate systems introduces distortions in the perception information, resulting in issues such as "Weak Adaptability to Heatmap Distribution"and "Offset in the Center Point of the Bounding Box."To address these challenges, this paper proposes a cutting-edge 3D object detection model named PolarBEVU, which leverages the bird's-eye view under the Polar coordinates along with multi-camera unprojection. The model introduces an innovative "Deformable Uniform Heatmap Distribution"method that adjusts heatmap computations based on box shapes, generating high-quality heatmaps and effectively resolving the issue of "Weak Adaptability to Heatmap Distribution."Moreover, the model incorporates the concept of "Dynamic High-risk Regression Region"to enhance the accuracy and robustness of the center point regression at the bounding box, thus mitigating the issue of "Offset in the Center Point of the Bounding Box."In extensive experiments on the nuScenes dataset, PolarBEVU achieves impressive results with 49.9% mAP and 57.4% NDS on the test set, surpassing other comparative approaches and reaching the state-of-the-art (SOTA) performance among methods utilizing Polar coordinates. This clearly demonstrates the efficacy and superiority of PolarBEVU. In addition, the model is successfully deployed on Nvidia Jetson AGX Orin, showcasing real-time inference speeds of 31.42ms. These findings affirm PolarBEVU's potential for practical applications. Code is available at https: //***/JLUrob/PolarBEVU. © 2025 IEEE.

关键词： Object tracking