检索结果-内蒙古大学图书馆

A New Benchmark and Algorithm for Clothes-Changing Video Person Re-Identification

IEEE Transactions on Information Forensics and Security 2025年 20卷 1993-2005页

作者： Wang, Likai Zhang, Xiangqun Han, Ruize Wei, Yanjie Wang, Song Feng, Wei Tianjin University College of Intelligence and Computing Tianjin300350 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518107 China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China

Person re-identification (Re-ID) is a classical computer vision task and has significant applications for public security and information forensics. Recently, long-term Re-ID with clothes-changing has attracted increasing attention. However, existing methods mainly focus on image-based setting, where richer temporal information is overlooked. In this paper, we focus on the relatively new yet practical problem of Clothes-Changing Video-based Re-ID (CCVReID), which is less studied. First, given the dataset shortage, we build two new benchmark datasets for CCVReID problem, including a large-scale synthetic video dataset and a real-world one, both containing human sequences with various clothing changes. Moreover, we systematically study this problem by simultaneously considering the classical appearance feature and temporal feature contained in the video. We develop a dual-branch fusion framework that makes use of the information from both clothes-aware appearance feature and clothes-free gait feature. For better information fusion, a confidence-guided re-ranking strategy is proposed to adaptively balance the weight of these two categories of features. We have released the benchmark and code proposed in this work to the public at https://***/kkw98/CCVReID. © 2025 IEEE.

关键词： Reamers

来源：评论

学校读者我要写书评

暂无评论

Where to Turn: Road Fork Detection in Sparse 3D Point Cloud 3rd

Where to Turn: Road Fork Detection in Sparse 3D Point Clou...

引用

3rd International Conference on Autonomous Unmanned Systems, ICAUS 2023

作者： Hu, Di Zhang, Kai Zhong, Yipan Xu, Jiachen Yuan, Xia Zhao, Chunxia School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China Research Institute of Intelligence Southwest Research Institute of Information Control Chengdu611700 China

ISBN: (纸本)9789819711024

Intersection detection plays a crucial role in localizing and planning the path of autonomous vehicles in urban environments. This paper presents a novel approach, PVWO, for adaptive intersection detection in autonomous vehicles equipped with 3D LiDAR. Firstly, the algorithm removes ground and obstacles effectively, and eliminates noise points. Secondly, a polyarticular viewpoints beam model is applied to detect the type and location of intersections. Thirdly, a linear feature extraction approach in a rolling window is proposed to detect road width and optimize key parameters, enhancing model robustness across different scenes. Compared to parallel intersection detection algorithms, our method exhibits excellent performance under diverse road conditions. Experiments in sparse point clouds show an average precision exceeding 89% and an average processing time of approximately 88 ms/frame. These results demonstrate that our method can accurately detect intersections in real-time, across various road structures. © Beijing HIWING Scientific and Technological Information institute 2024.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

A Novel Approximation Algorithm for Max-Covering Circle Problem 16th

A Novel Approximation Algorithm for Max-Covering Circle Pro...

引用

16th Annual International Conference on Combinatorial Optimization and Applications, COCOA 2023

作者： Zhang, Kaiqi Zhang, Siyuan Gao, Jirun Wang, Hongzhi Gao, Hong Li, Jianzhong School of Computer Science and Technology Harbin Institute of Technology Harbin China School of Computer Science and Technology Zhejiang Normal University Jinhua China Faculty of Computer Science and Control Engineering Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

ISBN: (纸本)9783031496103

We study the efficient approximation algorithm for max-covering circle problem. Given a set of weighted points in the plane and a circle with specified size, max-covering circle problem is to find the proper place where the center of the circle is located so that the total weight of the points covered by the circle is maximized. Our core approach is to approximate the circle with a symmetrical rectilinear polygon (SRP). We first present a method to construct the circumscribed SRP of a given circle and disclose their area relationship. Then, we convert max-covering SRP problem to SRP intersection problem, which can be efficiently solved with simple partition and modification based on the existing method. Finally, the optimal solution returned from max-covering SRP problem can be used to produce an approximate answer to max-covering circle problem. We prove that for most of the inputs, our algorithm can give a (1 - Ε) approximation to the optimal solution, which only needs O(nΕ-1logn+nΕ-1log(1Ε)) time for unit points and o(nΕ-2logn) time for weighted points. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Analysis of CNN and Feed-Forward ANN Model for the Evaluation of ECG Signal

引用

Current Signal Transduction Therapy 2023年第1期18卷 1-8页

作者： Mathur, Prerak Sharma, Tanu Veer, Karan Department of Instrumentation and Control Engineering Department of Computer Science and Engineering Dr. B. R. Ambedkar National Institute of Technology Punjab Jalandhar India

Background: Heart disease is considered one of the complex diseases that has affected a large number of people around the world. It is important to detect and identify cardiac diseases at early stages. Objective: A large number of methods are already present that detect various heart diseases;how-ever, there are some limitations to these methods, that have degraded their overall performance. Methods: In this paper, an effective and efficient method based on a convolutional neural network (CNN) and a feed-forward artificial neural network (FFANN) is proposed that can effectively detect cardiac diseases after analysing the electrocardiogram (ECG) signals. In this ongoing study, the transformed signals are used to extract the information from the processed data. The extracted features are then passed to the proposed CNN-FFANN classifiers for training and testing purposes. Results: The performance of the proposed CNN-FFANN model is evaluated in the MATLAB software in terms of performance matrices. Conclusion: The simulated outcomes have proved the proposed CNN-FFANN model to be more accurate and efficient in detecting heart diseases from ECG signals and can be adopted for future biomedical applications. © 2023 Bentham Science Publishers.

关键词： artificial intelligence artificial neural network biomedical applications cardiac diseases ECG signal health monitoring system

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Neural Network: Hidden Firing Patterns, Phase Synchronization and Multistability

Heterogeneous Neural Network: Hidden Firing Patterns, Phase ...

引用

2023 IEEE International Conference on Memristive Computing and Applications, ICMCA 2023

作者： Huang, Lili Zang, Hongyan Lei, Tengfei Li, Na Fu, Haiyan Liu, Aidi Jinan China School of Intelligent Manufacturing and Control Engineering Qilu Institute of Technology Jinan China School of Computer and Information Engineering Qilu Institute of Technology Jinan China

ISBN: (纸本)9798350316230

In this paper, a novel heterogeneous neural network is proposed by coupling improved memristor 2D HR neuron and 1D HNN neuron with locally active memristor. The effect of coupling intensity on synchronization behavior is analyzed. The results show that the increase in coupling intensity will lead to phase synchronization of the system. The coexistence of the firing pattern caused by the initial enhancement of the memristor is quantitatively analyzed and numerically simulated by introducing a multistable auto memristor. The results show that the newly constructed heterogeneous neural network exhibits an attractor bias pattern that enhances the initial migration behavior of the memristor, and these patterns have extreme multi-stability. © 2023 IEEE.

关键词： Memristors

来源：评论

学校读者我要写书评

暂无评论

Cloud-Edge Intelligent Collaborative Computing Model Based on Transfer Learning in IoT 9th

Cloud-Edge Intelligent Collaborative Computing Model Based o...

引用

9th International Conference of Pioneering computer Scientists, Engineers and Educators, ICPCSEE 2023

作者： Long, Yang Li, Zhixin School of Information and Control Engineering Jilin Institute of Chemical Technology Jilin132022 China School of Computer Technology and Engineering Changchun Institute of Technology Changchun130012 China

ISBN: (纸本)9789819959679

The deep neural network is a reliable technical support for cloud computing and edge computing. It has excellent nonlinear approximation and generalization capabilities, making it suitable for classifying and predicting Internet of Things data in cloud computing and edge computing fields. However, the increasing size of neural networks poses a challenge for their deployment on devices with limited computing and storage resources. Traditional cloud computing services also suffer from high latency, which hinders real-time tasks. To address these challenges, this paper proposes a cloud-side cooperation model for deep learning based on migration learning technology. This model used migration learning technology to reduce the size of deep neural networks. Specifically, it deployed the deep neural network model (CDLM) in the cloud and the shallow neural network model (EDLM) at the edge. CDLM is used to help train EDLM and improve its performance, enabling it to run independently on edge devices with high accuracy and respond to real-time tasks. This approach reduced the amount of user data transmitted to the cloud, alleviated bandwidth pressure, and protected user privacy. Experimental results show that the proposed model improved the accuracy of EDLM by 19.58% compared with traditional neural network models. These findings provide a theoretical and experimental foundation for the study of cloud-edge collaborative models. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

Emotion detection on phone calls during emergency using ensemble model with hyper parameter tuning

引用

International Journal of Information Technology (Singapore) 2023年第2期15卷 745-757页

作者： Selvan, A. Kalai Nimmi, K. Janet, B. Sivakumaran, N. C-DAC Thiruvananthapuram India Department of Computer Applications National Institute of Technology Tiruchirappalli India Department of Instrumentation and Control Engineering National Institute of Technology Tiruchirappalli India

The paper’s main focus is on speech-based emotion detection in Malayalam phone call records, including emergency calls [(emergency response support system (ERSS)] and elicited *** emotions taken into consideration are fear, anger, sad, disgust, panic, anxious are among the emotions considered, while all other emotions are labeled as others. When reporting an emergency, callers commonly exhibit emotional stress, which results in speech output discrepancies. In such uncontrolled situations, determining caller qualities is exceedingly challenging due to poor speech quality and background noise. Fear, anger, and sadness are the three basic emotions expressed during an emergency. In order to classify the emotions, this paper used a bagging and boosting ensemble model. Using the Malayalam speech emotion dataset (MSED), the proposed support vector machine ensembles with Bagging with radial basis function kernel function were found to have an accuracy of 93.01%. Elicited and emergency speech from emergency calls are used to train support vector machine ensembles. For comparison of the performance of the model, we have used RAVDESS and Emo-DB. The accuracy obtained were 81.40% while using the RAVDESS dataset and 93.01% while using Emo-DB dataset which is high when compared to existing machine learning and ensemble model. The performance of boosting models was comparatively lesser than that of the Bagging model. © 2022, The Author(s), under exclusive licence to Bharati Vidyapeeth's institute of computer Applications and Management.

关键词： Emergency response support system (ERSS) Ensemble model Mel frequency cepstral co-efficient (MFCC) Speech emotion detection (SED) Support vector machine (SVM) ZeroCrossing rate (ZCR)

来源：评论

学校读者我要写书评

暂无评论

Fast Finite-Time Adaptive Neural control for Pure-Feedback Stochastic Nonlinear Systems

Fast Finite-Time Adaptive Neural Control for Pure-Feedback S...

引用

2023 IEEE International Conference on Memristive Computing and Applications, ICMCA 2023

作者： Zhao, Ximei Zhang, Hanyu Shao, Lijun Huang, Lili Li, Jiao School of Intelligent Manufacturing and Control Engineering Qilu Institute of Technology Jinan China Qilu Institute of Technology Jinan China College of Computer and Communication Engineering Qilu Institute of Technology Jinan China

ISBN: (纸本)9798350316230

In this article, we solve the fast finite-time stabilization as well as adaptive neural control design issues for a class uncertain stochastic nonlinear systems. By employing the mean value theorem, the pure-feedback form can be transformed into strict-feedback case. Firstly, a new fast finite-time control scheme is proposed in virtue of generalizing the fast finite-time adaptive control strategy for deterministic systems to the stochastic case. Next, a novel adaptive neural control strategy is developed to simultaneously deal with the stochastic nonlinear systems with completely unknown nonlinearities as well as the disturbances term. Then, stability analysis have been given based on a Jensen's inequality. © 2023 IEEE.

关键词： Feedback

来源：评论

学校读者我要写书评

暂无评论

Dynamic Modeling of Key Operating Parameters for Supercritical Circulating Fluidized Bed Units based on Data-Knowledge-Driven Method

引用

Journal of Thermal Science 2024年第3期33卷 1216-1230页

作者： YU Haoyang GAO Mingming ZHANG Hongfu CHEN Jiyu LIU Jizhen YUE Guangxi The State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources School of Control and Computer EngineeringNorth China Electric Power UniversityBeijing102206China CHN Energy New Energy Technology Research Institute Co. Ltd.Beijing102209China Department of Energy and Power Engineering Tsinghua UniversityBeijing100084China

To address the pressing need for intelligent and efficient control of circulating fluidized bed(CFB)units,it is crucial to develop a dynamic model for the key operating parameters of supercritical circulating fluidized bed(SCFB)***,data-knowledge-driven dynamic model of bed temperature,load,and main steam pressure of the SCFB unit has been ***,a knowledge-driven method is employed to develop a dynamic model for key operating parameters of SCFB *** model parameters are determined based on the operating data of the unit and continuously optimized in real ***,Bidirectional Long Short-Term Memory combined with Convolutional Neural Network and Attention Mechanism is utilized to build the dynamic model of bed temperature,load,and main steam ***,a collaboration and integration method based on the critic weight method and the variation coefficient method is proposed to establish data-knowledge-driven model of key operating parameters for SCFB *** model displays great accuracy and fitting ability compared with other methods and effectively captures the dynamic characteristics,which can provide a research basis for the design of intelligent flexible control mode of SCFB unit.

关键词： circulating fluidized bed dynamic model data-knowledge-driven method deep learning power plant

来源：评论

学校读者我要写书评

暂无评论

Binary Stochastic Flip Optimization for Training Binary Neural Networks

Binary Stochastic Flip Optimization for Training Binary Neur...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Shibuya, Tatsukichi Inoue, Nakamasa Kawakami, Rei Sato, Ikuro Department of Computer Science Institute of Science Tokyo Tokyo Japan Department of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan Denso IT Laboratory Tokyo Japan

ISBN: (纸本)9798350368741

For deploying deep neural networks on edge devices with limited resources, binary neural networks (BNNs) have attracted significant attention, due to their computational and memory efficiency. However, once a neural network is binarized, finetuning it on edge devices becomes challenging because most conventional training algorithms for BNNs are designed for use on centralized servers and require storing real-valued parameters during training. To address this limitation, this paper introduces binary stochastic flip optimization (BinSFO), a novel training algorithm for BNNs. BinSFO employs a parameter update rule based on Boolean operations, eliminating the need to store real-valued parameters and thereby reducing memory requirements and computational overhead. In experiments, we demonstrated the effectiveness and memory efficiency of BinSFO in finetuning scenarios on six image classification datasets. BinSFO performed comparably to conventional training algorithms with a 70.7% smaller memory requirement. Code is released at https://***/TatsukichiShibuya/ICASSP2025_BinSFO. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：