检索结果-内蒙古大学图书馆

The FaceChannel: A Fast and Furious Deep Neural Network for Facial Expression Recognition

SN computer Science 2020年第6期1卷 321页

作者： Barros, Pablo Churamani, Nikhil Sciutti, Alessandra Cognitive Architecture for Collaborative Technologies Unit Istituto Italiano di Tecnologia Genoa Italy Department of Computer Science and Technology University of Cambridge Cambridge United Kingdom

Current state-of-the-art models for automatic facial expression recognition (FER) are based on very deep neural networks that are effective but rather expensive to train. Given the dynamic conditions of FER, this characteristic hinders such models of been used as a general affect recognition. In this paper, we address this problem by formalizing the FaceChannel, a light-weight neural network that has much fewer parameters than common deep neural networks. We introduce an inhibitory layer that helps to shape the learning of facial features in the last layer of the network and, thus, improving performance while reducing the number of trainable parameters. To evaluate our model, we perform a series of experiments on different benchmark datasets and demonstrate how the FaceChannel achieves a comparable, if not better, performance to the current state-of-the-art in FER. Our experiments include cross-dataset analysis, to estimate how our model behaves on different affective recognition conditions. We conclude our paper with an analysis of how FaceChannel learns and adapts the learned facial features towards the different datasets. © 2020, The Author(s).

关键词： Affective computing Convolutional neural network Facial expression recognition

来源：评论

学校读者我要写书评

暂无评论

A Shared Pose Regression Network for Pose Estimation of Objects from RGB Images

A Shared Pose Regression Network for Pose Estimation of Obje...

引用

International IEEE Conference on Signal-Image Technologies and Internet-Based System

作者： Stefan Hein Bengtson Hampus Åström Thomas B. Moeslund Elin A. Topp Volker Krueger Department of Architecture Design and Media Technology Visual Analysis and Perception (VAP) Laboratory Pioneer Centre for AI Denmark Aalborg University Aalborg Denmark Department of Computer Science Lund University Robotics and Semantic Systems Lund Sweden

In this paper we propose a shared regression network to jointly estimate the pose of multiple objects, replacing multiple object-specific solutions. We demonstrate that this shared network can outperform other similar approaches that rely on multiple object-specific models by evaluating it on the TLESS dataset using the VSD (Visible Surface Discrepancy). Our approach offers a less complex solution, with fewer parameters, lower memory consumption and less training required. Furthermore, it inherently handles symmetric objects by using a depth-based loss during training and can predict in real-time. Finally, we show how our proposed pipeline can be used for fine-tuning a feature extractor jointly on all objects while training the shared pose regression network. This fine-tuning process improves the pose estimation performance.

关键词： Training Pose estimation Memory management Pipelines Feature extraction Rendering (computer graphics) Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Reliable Multihop Routing and Congestion Aware Chicken Swarm Optimization in Vehicular Ad-Hoc Network 3

Reliable Multihop Routing and Congestion Aware Chicken Swarm...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Anitha, Cuddapah Al-Tarazi, Dalia Vignesh, G.D. Vasuki, A. Vinotha Giri, Jayant Dinesh, M. School of Computing Mohan Babu University Erstwhile Sree Vidyanikethan Engineering College Department of Computer Science and Engineering Tirupati India Zarqa University Faculty of Engineering Department of Architecture Engineering Zarqa Jordan College of Engineering University of Business and Technology Jeddah Saudi Arabia St. Joseph's College of Engineering Department of Electronics and Communication Engineering Chennai India J.P. College of Engineering Department of Computer Science and Engineering Tenkasi India Yeshwantrao Chavan College of Engineering Department of Mechanical Engineering Nagpur India Lovely Professional University Division of Research and Development Phagwara India Department of Vlsi Microelectronics Chennai India

ISBN: (纸本)9798331508456

Background: Vehicular Ad Hoc Networks (VANETs) play a crucial role in intelligent transportation by facilitating communication between vehicles and vehicles with infrastructure-based models. They encounter problems such as significant movement, overcrowding, and loss of data. This research presents a novel deep reinforcement learning-based resource allocation and congestion optimization (DRLRCO) framework aimed at improving communication efficiency and high reliability using an effective learning process. Method Used: The DRLRCO framework employs agent-based learning to ensure precise data transmission, AdMAC Protocol for dynamic packet delivery, and Congestion-Aware Chicken Swarm Optimization (CSO) to enhance network performance. Result: The DRLRCO framework surpasses reinforcement learning models for comprehensive testing. Significant enhancements involve improved accuracy, reduced data loss, minimized overhead, increased throughput, and shorter average delays for real-time communication. Conclusion: The DRLRCO framework successfully tackles congestion and data loss in vehicular communication. Future studies might involve Artificial Intelligence (AI) for traffic forecasting, experimentation among high-speed vehicles and investigating blockchain for enhanced security. © 2025 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Discrete Jaya algorithm to solve the police resource deployment problem in emergency traffic accident 10

Discrete Jaya algorithm to solve the police resource deploym...

引用

10th IEEE Joint International Information technology and Artificial Intelligence Conference, ITAIC 2022

作者： Jing, Zhang Weijie, Ding Zhejiang Police College Department of Computer and Information Security Hangzhou310053 China Big-data and Network Security Research Institute Zhejiang Police College Hangzhou310053 China College of Information Engineering Zhejiang University of Technology Hangzhou310023 China Key Laboratory of Public Security Information Application Based on Big-data Architecture Ministry of Public Security Hangzhou310053 China

ISBN: (数字)9781665422079

ISBN: (纸本)9781665422079

Aiming at the high frequency of urban traffic accidents and the relative shortage of police resources, a multi-objective discrete Jaya algorithm is proposed to solve the problem of police resource deployment in emergency traffic accident handling. The purpose is to make timely response and handling of traffic accidents through the rational allocation of police deployment points in the jurisdiction. Taking the coverage of accident points and the average response time of accidents as the optimization objectives, the discrete Jaya algorithm is used as the global search strategy to approach the Pareto front by constantly approaching the optimal solution and far away from the worst solution, and then the neighborhood search strategy is used to improve the local search ability of the algorithm. Finally, a practical example is given to verify the effectiveness of the proposed method. © 2022 IEEE.

关键词： Law enforcement

来源：评论

学校读者我要写书评

暂无评论

IEEE 8th VR Workshop on Sonic Interactions for Virtual Environments

IEEE 8th VR Workshop on Sonic Interactions for Virtual Envir...

引用

Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW), IEEE Conference on

作者： Stefania Serafin Ali Adjorlu Rolf Nordahl Federico Fontana Michele Geronazzo Tifanie Bouchara Florent Berthaut Romain Michon Dorte Hammershøj Lorenzo Picinali Department of Architecture Design and Media Technology Aalborg University Copenhagen Denmark Department of Mathematics Computer Science and Physics University of Udine Udine Italy Department of Engineering and Management University of Padova Vicenza Italy LISN CNRS Université Paris-Saclay France CRIStAL CNRS Univ. Lille Lille France INRIA Lyon France Department of Electronic Systems Aalborg University Denmark Dyson School of Design Engineering Imperial College London London UK

ISBN: (数字)9798331514846

ISBN: (纸本)9798331525637

Sonic interaction design is defined as the study and exploitation of sound as one of the principal channels conveying information, meaning, and aesthetic/emotional qualities in interactive contexts. This field lies at the intersection of interaction design, and sound and music *** the virtual reality community, the focus of research on topics related to auditory feedback has been rather limited when compared, for example, to the focus placed on visual feedback or even on haptic feedback. However, in such communities as film production or product sound design it is well known that sound is a powerful way to communicate meaning and emotion to a scene or a *** 2025 is the 8th of a series of workshops, whose main goal is to increase awareness among the virtual reality community of the importance of sonic elements when designing virtual/augmented/mixed reality (XR hereafter) environments. Participants to the workshop will also discuss how research in other related fields such as film sound theory, product sound design, sound and music computing, game sound design, and accessibility can inform designers of XR environments. Moreover, the workshop will feature state-of-the-art research in the field of sound for XR environments.

关键词： Visualization Three-dimensional displays Conferences Spatial audio Music Virtual environments Production Games User interfaces Haptic interfaces

来源：评论

学校读者我要写书评

暂无评论

Tropes in films: An initial analysis

arXiv

引用

arXiv 2020年

作者： García-Ortega, Rubén Héctor García-Sánchez, Pablo Merelo-Guervós, Juan Julián Department of Computer Architecture and Technology University of Granada Department of Languages and Computer Systems University of Granada Department of Computer Architecture and Technology University of Granada

TVTropes is a wiki that describes tropes and which ones are used in which artistic work. We are mostly interested in films, so after releasing the TropeScraper Python module that extracts data from this site, in this report we use scraped information to describe statistically how tropes and films are related to each other and how these relations evolve in time. In order to do so, we generated a dataset through the tool TropeScraper in April 2020. We have compared it to the latest snapshot of DB Tropes, a dataset covering the same site and published in July 2016, providing descriptive analysis, studying the fundamental differences and addressing the evolution of the wiki in terms of the number of tropes, the number of films and connections. The results show that the number of tropes and films doubled their value and quadrupled their relations, and films are, at large, better described in terms of tropes. However, while the types of films with the most tropes has not changed significantly in years, the list of most popular tropes has. This outcome can help on shedding some light on how popular tropes evolve, which ones become more popular or fade away, and in general how a set of tropes represents a film and might be a key to its success. The dataset generated, the information extracted, and the summaries provided are useful resources for any research involving films and tropes. They can provide proper context and explanations about the behaviour of models built on top of the dataset, including the generation of new content or its use in machine learning. © 2020, CC BY-SA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Drive-Awake: A YOLOv3 Machine Vision Inference Approach of Eyes Closure for Drowsy Driving Detection 3

Drive-Awake: A YOLOv3 Machine Vision Inference Approach of E...

引用

3rd IEEE International Conference on Artificial Intelligence in Engineering and technology, IICAIET 2021

作者： MacAlisang, Jonel R. Alon, Alvin Sarraga Jardiniano, Moises F. Evangelista, Deanne Cameren P. Castro, Julius C. Tria, Meriam L. Technology Licensing Office-Itso Technological University of the Philippines Manila Philippines Digital Transformation Center Steer Hub Batangas State University Batangas City Philippines Feu Institute of Technology Department of Computer Engineering Manila Philippines College of Engineering and Architecture Bohol Island State University Tagbilaran Bohol Philippines Eulogio Amang Rodriguez Institute of Science and Technology Department of Electronics Engineering Manila Philippines

ISBN: (纸本)9781665428996

Nowadays, road accidents have become a major concern. The drowsiness of drivers owing to overfatigue or tiredness, driving while intoxicated, or driving too quickly is some of the primary causes of this. Drowsy driving contributes to or increases the number of traffic accidents each year. The study presented a technique for detecting driver drowsiness in response to this issue. The sleep states of the drivers in the driving environment were detected using a deep learning approach. To assess if the eyes of particular constant face images of drivers are closed, a convolutional neural network (CNN) model has been developed. The suggested model has a wide range of possible applications, including human-computer interface design, facial expression detection, and determining driver tiredness and drowsiness. The YOLOv3 algorithm, as well as additional tools like Pascal VOC and LabelImg, were used to build this approach, which collects and trains a driver dataset that feels drowsy. The study's total detection accuracy was 100%, with detection per frame accuracy ranging from 49% to 89%. © 2021 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Deep learning based Identification of Solid Waste Management in Smart Cities through Garbage Separation and Monitoring 5

Deep learning based Identification of Solid Waste Management...

引用

5th International Conference on Inventive Research in Computing Applications, ICIRCA 2023

作者： George, Abin Mankar, Sham H. Suryanarayana, N.V.S. Gupta, Manish Sathish, R. Priya, A. Knowledge and Technology Management Division Council of Scientific & Industrial Research National Aerospace Laboratories Karnataka Bangaluru India Pimpri Chinchwad College of Engineering and Research Department of Mechanical Engineering Maharashtra Pune India Central Tribal University of Andhra Pradesh Administrative Officer Andhra Pradesh Vizianagaram India Lovely Professional University Division of Research and Development Punjab Phagwara India Shri Nehru Maha Vidayala College of Arts and Science Department of Computer Science Tamil Nadu Coimbatore India School of Architecture Koneru Lakshmaiah Educational Foundation Andhra Pradesh Gundur India

ISBN: (纸本)9798350321425

The solid waste management is the process of proper decomposition of waste materials within a period of time. This includes the collection of garbage's and then proceeded through certain measures for decomposition. There are various methods adopted in the garbage separation process. This includes the artificial intelligence techniques for the estimation and determination of the solid waste through automatic detection and separation of the garbage waste using control and sensing units. They are integrated with the internet of things to enable the two way communication system. This helps to visualize the functioning of the system adopting the digital platform. The proposed system is implemented through the smart dust bin held in every household that automatically senses the non-biodegradable and biodegradable waste materials. The classification of the waste materials are identified through the image processing techniques. © 2023 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Edge Computing in Smart Cities: Enhancing Real-Time Data Processing

Edge Computing in Smart Cities: Enhancing Real-Time Data Pro...

引用

Advances in Computation, Communication and Information technology (ICAICCIT), International Conference on

作者： S Rajakumari C Natarajan Vinod N. Alone P Senthilkumar R. Gayathri J Godwin John Department of Mathematics R.M.D. Engineering College Chennai India Department of CSE P. S. R Engineering College Sivakasi Tamilnadu India Computer Engineering Department VPPCOE &VA Sion Mumbai Maharashtra India School of Architecture Koneru Lakshmaiah Education Foundation Vaddeswaram Guntur AP India Department of ECE Rajalakshmi Engineering College Chennai Tamilnadu India Department of Mechanical Engineering Rajalakshmi Institute of Technology Tamil Nadu India

ISBN: (数字)9798331541217

ISBN: (纸本)9798331541224

One of the most crucial issues that smart cities have to keep addressing as they grow is urban mobility improvement. Maximizing the efficiency of transportation, so lowering congestion, and so optimizing traffic flow can help to solve the growing urbanization and increasing vehicle numbers. This research aims to develop and implement a deep learning method based on DenseNet in order to reach the target of raising the effectiveness of urban traffic management. The unique selling proposition that distinguishes this sector from others is the introduction of modern technologies into a sector that has always depended on conventional methods. The main goal of the work was to apply DenseNet architecture to analyze traffic patterns. Considering the two other strategies, the proposed option has a lot of possibilities. From evaluation criteria-which comprised recall, precision, and F1-score, the proposed solution outperformed the approaches regarded to be state-of-the-art.

关键词： Deep learning Smart cities Transportation computer architecture Traffic control Market research Data processing Real-time systems Information and communication technology Edge computing

来源：评论

学校读者我要写书评

暂无评论

Adaptable Invisibility Management Using Kirigami-Inspired Transformable Metamaterials

引用

Research 2021年第1期2021卷 1034-1044页

作者： He-Xiu Xu Mingzhao Wang Guangwei Hu Shaojie Wang Yanzhao Wang Chaohui Wang Yixuan Zeng Jiafang Li Shuang Zhang Wei Huang Air and Missile Defense College Air Force Engineering UniversityXi’an 710051China Institute of Flexible Electronics Northwestern Polytechnical UniversityXi’an 710072China Department of Electrical and Computer Engineering National University of SingaporeSingaporeSingapore 117583 Centre for Quantum Physics Key Laboratory of Advanced Optoelectronic Quantum Architecture and Measurement(MOE)Beijing 100081China School of Physics Beijing Institute of TechnologyBeijing 100081China Department of Physics University of Hong KongHong KongChina Department of Electronic&Electrical Engineering University of Hong KongHong KongChina

Many real-world applications,including adaptive radar scanning and smart stealth,require reconfigurable multifunctional devices to simultaneously manipulate multiple degrees of freedom of electromagnetic(EM)waves in an on-demand ***,kirigami technique,affording versatile and unconventional structural transformation,has been introduced to endow metamaterials with the capability of controlling EM waves in a reconfigurable ***,we report for a kirigami-inspired sparse meta-architecture,with structural density of 1.5%in terms of the occupation space,for adaptive invisibility based on independent operations of frequency,bandwidth,and *** on the general principle of dipolar management via structural reconstruction of kirigami-inspired meta-architectures,we demonstrate reconfigurable invisibility management with abundant EM functions and a wide tuning range using three enantiomers(A,B,and C)of different geometries characterized by the folding angleβ.Our strategy circumvents issues of limited abilities,narrow tuning range,extreme condition,and high cost raised by available reconfigurable metamaterials,providing a new avenue toward multifunctional smart devices.

关键词： tuning visibility smart

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：