检索结果-南通市图书馆

ResLNet:deep residual LSTM network with longer input for action recognition

Frontiers of Computer Science 2022年第6期16卷 41-49页

作者： Tian WANG Jiakun LI Huai-Ning WU Ce LI hichem snoussi Yang WU Institute of Artificial Intelligence Beihang UniversityBeijing100191China School of Automation Science and Electrical Engineering Beihang UniversityBeijing100191China College of Electrical and Information Engineering Lanzhou University of TechnologyLanzhou730050China Institute Charles Delaunay-LM2S FRE CNRS 2019 University of Technology of TroyesTroyes10010France Institute for Research Initiatives Nara Institute of Science and TechnologyNara630-0192Japan

Action recognition is an important research topic in video analysis that remains very challenging.Effective recognition relies on learning a good representation of both spatial information(for appearance)and temporal information(for motion).These two kinds of information are highly correlated but have quite different properties,leading to unsatisfying results of both connecting independent models(e.g.,CNN-LSTM)and direct unbiased co-modeling(e.g.,3DCNN).Besides,a long-lasting tradition on this task with deep learning models is to just use 8 or 16 consecutive frames as input,making it hard to extract discriminative motion features.In this work,we propose a novel network structure called ResLNet(Deep Residual LSTM network),which can take longer inputs(e.g.,of 64 frames)and have convolutions collaborate with LSTM more effectively under the residual structure to learn better spatial-temporal representations than ever without the cost of extra computations with the proposed embedded variable stride convolution.The superiority of this proposal and its ablation study are shown on the three most popular benchmark datasets:Kinetics,HMDB51,and UCF101.The proposed network could be adopted for various features,such as RGB and optical flow.Due to the limitation of the computation power of our experiment equipment and the real-time requirement,the proposed network is tested on the RGB only and shows great performance.

关键词： action recognition deep learning neural network

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Accelerating temporal action proposal generation via high performance computing

引用

Frontiers of Computer Science 2022年第4期16卷 59-68页

作者： Tian Wang Shiye Lei Youyou Jiang Choi Chang hichem snoussi Guangcun Shan Yao Fu Institute of Artificial Intelligence Beihang UniversityBeijing100191China School of Automation Science and Electrical Engineering Beihang UniversityBeijing100191China School of Software Tsinghua UniversityBeijing100084China Department of Computer Engineering Gachon UniversitySeongnam13120South Korea Institute Charles Delaunay-LM2S FRE CNRS 2019 University of Technology of TroyesTroyes10010France School of Instrumentation Science and Opto-electronics Engineering Beihang UniversityBeijing100191China Changchun Institute of Optics Fine Mechanics and PhysicsChinese Academy of SciencesChangchun130033China

Temporal action proposal generation aims to output the starting and ending times of each potential action for long videos and often suffers from high computation cost.To address the issue,we propose a new temporal convolution network called Multipath Temporal ConvNet(MTCN).In our work,one novel high performance ring parallel architecture based is further introduced into temporal action proposal generation in order to respond to the requirements of large memory occupation and a large number of videos.Remarkably,the total data transmission is reduced by adding a connection between multiple-computing load in the newly developed architecture.Compared to the traditional Parameter Server architecture,our parallel architecture has higher efficiency on temporal action detection tasks with multiple GPUs.We conduct experiments on ActivityNet-1.3 and THUMOS14,where our method outperforms-other state-of-art temporal action detection methods with high recall and high temporal precision.In addition,a time metric is further proposed here to evaluate the speed performancein the distributed training process.

关键词： temporal convolution temporal action proposal generation deep learning

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Abnormal event detection via the analysis of multi-frame optical flow information

引用

Frontiers of Computer Science 2020年第2期14卷 304-313页

作者： Tian WANG Meina QIAO Aichun ZHU Guangcun SHAN hichem snoussi School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100191China School of Computer Science and Technology Nanjing University of TechnologyNanjing 210094China School of Instrumentation Science and Opto-electronics Engineering Beihang UniversityBeijing 100191China Institute Charles Delaunay-LM2S-UMR STMR 6281 CNRS University of Technology of TroyesTroyes 10010France

Security surveillance of public scene is closely relevant to routine safety of individual.Under the stimulus of this concern,abnormal event detection is becoming one of the most important tasks in computer vision and video processing.In this paper,we propose a new algorithm to address the visual abnormal detection problem.Our algorithm decouples the problem into a feature descriptor extraction process,followed by an AutoEncoder based network called cascade deep AutoEncoder(CDA).The movement information is represented by a novel descriptor capturing the multi-frame optical flow information.And then,the feature descriptor of the normal samples is fed into the CDA network for training.Finally,the abnormal samples are distinguished by the reconstruction error of the CDA in the testing procedure.We validate the proposed method on several video surveillance datasets.

关键词： abnormal event detection multi-frame optical flow cascade deep autoencoder

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Crysformer:An attention-based graph neural network for properties prediction of crystals

引用

中国物理B:英文版 2023年第9期32卷 15-20页

作者：王田陈家辉滕婧史金钢曾新华 hichem snoussi Institute of Artificial Intelligence SKLSDEBeihang UniversityBeijing 100191China Zhongguancun Laboratory Beijing 100191China School of Aeronautic Science and Engineering Beihang UniversityBeijing 100191China School of Control and Computer Engineering North China Electric Power UniversityBeijing 102206China School of Software Engineering Xi'an Jiaotong UniversityXi'an 710049China Academy for Engineering and Technology Fudan UniversityShanghai 200433China Charles Delaunay Institute University of Technology of TroyesTroyes Cedex 10004France

We present a novel approach for the prediction of crystal material properties that is distinct from the computationally complex and expensive density functional theory(DFT)-based calculations.Instead,we utilize an attention-based graph neural network that yields high-accuracy predictions.Our approach employs two attention mechanisms that allow for message passing on the crystal graphs,which in turn enable the model to selectively attend to pertinent atoms and their local environments,thereby improving performance.We conduct comprehensive experiments to validate our approach,which demonstrates that our method surpasses existing methods in terms of predictive accuracy.Our results suggest that deep learning,particularly attention-based networks,holds significant promise for predicting crystal material properties,with implications for material discovery and the refined intelligent systems.

关键词： deep learning property prediction crystal attention networks

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

AED-Net:An Abnormal Event Detection Network

引用

Engineering 2019年第5期5卷 930-939页

作者： Tian Wang Zichen Miao Yuxin Chen Yi Zhou Guangcun Shan hichem snoussi School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100191China Department of Electronic Engineering Dalian Maritime UniversityDalian 116026China School of Instrumentation Science and Opto-electronic Engineering&International Research Institute for Multidisciplinary Science Beihang UniversityBeijing 100191China Institute Charles Delaunay-LM2S-UMR STMR 6281 CNRS University of Technology of TroyesTroyes 10010France

It has long been a challenging task to detect an anomaly in a crowded scene.In this paper,a selfsupervised framework called the abnormal event detection network(AED-Net),which is composed of a principal component analysis network(PCAnet)and kernel principal component analysis(kPCA),is proposed to address this problem.Using surveillance video sequences of different scenes as raw data,the PCAnet is trained to extract high-level semantics of the crowd’s situation.Next,kPCA,a one-class classifier,is trained to identify anomalies within the scene.In contrast to some prevailing deep learning methods,this framework is completely self-supervised because it utilizes only video sequences of a normal situation.Experiments in global and local abnormal event detection are carried out on Monitoring Human Activity dataset from University of Minnesota(UMN dataset)and Anomaly Detection dataset from University of California,San Diego(UCSD dataset),and competitive results that yield a better equal error rate(EER)and area under curve(AUC)than other state-of-the-art methods are observed.Furthermore,by adding a local response normalization(LRN)layer,we propose an improvement to the original AED-Net.The results demonstrate that this proposed version performs better by promoting the framework’s generalization capacity.

关键词： Abnormal events detection Abnormal event detection network Principal component analysis network Kernel principal component analysis

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Multi-objective Optimization for Target Tracking in Quantized Sensor Networks

引用

通讯和计算机（中英文版） 2012年第10期9卷 1195-1205页

作者： Majdi Mansouri Faicel Hnaien Hazem Nounou Mohamed Nounou hichem snoussi Cedric Richard College of Engineering-Electrical and Computer Engineering Texas A &M University Doha 3874 Qatar ICD/LOSI University of Technology of Troyes Troyes l O000 France ICD/LM2S University of Technology of Troyes 10000 Troyes France Laboratoire FIZEA U UMR CNRS 6525 Nice Sophia Antipolis University 06000 Nice France

This work presents a multi-objective algorithm for jointly selecting the appropriate group of candidate sensors and optimizing the quantization for target tracking in WSN (wireless sensor networks). The authors focus on a more challenging problem of how to effectively utilize quantized sensor measurement for target tracking in sensor networks by considering sensors selection problem. Firstly, The authors jointly optimize the quantization level and the group of candidate sensors selection in order to provide the required data of the target and to balance the energy dissipation in the WSN. Then, The authors estimate the target position using QVF (quantized variational filtering) algorithm. The quantization optimization and the sensor selection are based on MO (multi-objective) that defines the main parameters that may influence the relevance of the participation in cooperation for target tracking. This selection is also based on the transmitting power between one sensor and the CH (cluster head). The best sensors selection and quantization optimization are designed to reduce the communication cost and the estimation error, which leads to a significant reduction of energy consumption and an accurate target tracking. The computation of these criteria is based on the target position predictive distribution provided by the QVF algorithm. The simulation results show that the proposed method outperforms the quantized variational filtering algorithm under sensing range constraint and the binary variational filtering algorithm.

关键词：无线传感器网络多目标优化目标跟踪量化滤波算法目标位置估计误差选择问题

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Using Gabor Filter in 3D Convolutional Neural Networks for Human Action Recognition

Using Gabor Filter in 3D Convolutional Neural Networks for H...

引用

第36届中国控制会议

作者： Jiakun Li Tian Wang Yi Zhou Ziyu Wang hichem snoussi School of Automation Science and Electrical Engineering Beihang University College of Information Science and Technology Dalian Maritime University

ISBN: (纸本)9781538629185

Human action recognition is an important topic in the field of computer vision.We use Gabor filter in 3D CNNs models in recognizing action.Convolutional neural networks（CNNs） are a type of deep learning models,which is an efficient recognition model and has a unique superiority in image processing.Three dimension convolutional neural networks can well analyze action from video data.Gabor filter is a special convolution kernel.Its performance in feature extraction is outstanding.We test out model by KTH dataset and achieve a well result.

关键词： Action recognition 3D CNNs Gabor filter

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Few-sample Multi-organ Abdominal Image Segmentation with Mean Teacher Model

Few-sample Multi-organ Abdominal Image Segmentation with Mea...

引用

第三十九届中国控制会议

作者： Pingyi Chen Tianyu Chen Zhiqiang Yang Tian Wang Mengyi Zhang hichem snoussi School of Automation Science and Electrical Engineering Beihang University Honor School Beihang University School of Electronic and Information Engineering Beihang University College of Electrical Engineering and Control Science Nanjing Tech University Institute Charles Delaunay-LM2S FRE CNRS 2019 University of Technology of Troyes

Medical segmentation is a significant task since it provides valuable information for diagnosis. In the recent years, convolutional neural networks have achieved great success in this field. However, the number of medical images is often limited which cannot support large networks to be trained. Thus, overfitting is much more common in medical images than other tasks. Also, to get annotated images is very costly and time-consuming. We stimulate an extreme condition where the number of the sample is so limited that we adopt a Mean Teacher Model to avoid overfitting. We build two models – student and teacher, with same structure and alternatively trained. We apply consistency loss to update the parameters of student and use Exponential Moving Average to compute parameters of teacher from student model. All code can be found in https://***/cpystan/MT-Model.

关键词： Image segmentation semi-supervised learning consistency loss

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Neural Style Transfer for Picture with Gradient Gram Matrix Description

Neural Style Transfer for Picture with Gradient Gram Matrix ...

引用

第三十九届中国控制会议

作者： Heng Jin Tian Wang Mengyi Zhang Mingmin Li Yan Wang hichem snoussi School of Automation Science and Electrical Engineering Beihang University College ofElectrical Engineering and Control Science Nanjing Tech University Internet of Things Technology Application Institute CASC Institue Charles Delaunay-LM2S FRE CNRS 2019University of Technology of Troyes

Despite the high performance of neural style transfer on stylized pictures,we found that Gatys et al. [1]algorithm cannot perfectly reconstruct texture style.Output stylized picture could emerge unsatisfied unexpected textures such like muddiness in local area and insufficient grain expression.Our method bases on original algorithm,adding the Gradient Gram description on style loss,aiming to strengthen texture expression and eliminate muddiness.To some extent our method lengthens the runtime,however,its output stylized pictures get higher perform ance on texture details,especially in the elimination of muddiness.

关键词： art style transfer neural style transfer picture stylize Gradient Gram matrix

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Abnormal Event Detection based on Deep Autoencoder fusing optical flow

Abnormal Event Detection based on Deep Autoencoder fusing op...

引用

第36届中国控制会议

作者： Meina Qiao Tian Wang Jiakun Li Ce Li Zhiwei Lin hichem snoussi School of Automation Science and Electrical Engineering Beihang University College of Electrical and Information Engineering Lanzhou University of Technology School of Computing Ulster University Institute Charles Delaunay-LM2S-UMR STMR 6279 CNRS University of Technology of Troyes

ISBN: (纸本)9781538629185

As an important research topic in computer vision,abnormal detection has gained more and more attention.In order to detect abnormal events effectively,we propose a novel method using optical flow and deep autoencoder.In our model,optical flow of the original video sequence is calculated and visualized as optical flow image,which is then fed into a deep autoencoder.Then the deep autoencoder extract features from the training samples which are compressed to low dimension vectors.Finally,the normal and abnormal samples gather separately in the coordinate axis.In the evaluation,we show that our approach outperforms the existing methods in different scenes,in terms of accuracy.

关键词： Abnormal event detection Deep autoencoder Optical flow

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：