检索结果-内蒙古大学图书馆

AVTENet: A Human-Cognition-Inspired Audio-Visual Transformer-Based Ensemble Network for Video Deepfake Detection

IEEE Transactions on Cognitive and Developmental Systems 2025年

作者： Hashmi, Ammarah Shahzad, Sahibzada Adil Lin, Chia-Wen Tsao, Yu Wang, Hsin-Min Taiwan International Graduate Program Academia Sinica Social Networks and Human-Centered Computing Program Taipei11529 Taiwan National Tsing Hua University Institute of Information Systems and Applications Hsinchu30013 Taiwan National Chengchi University Department of Computer Science Taipei11605 Taiwan National Tsing Hua University Department of Electrical Engineering Institute of Communications Engineering Hsinchu300044 Taiwan Academia Sinica Research Center for Information Technology Innovation Taipei11529 Taiwan

The recent proliferation of hyper-realistic deepfake videos has drawn attention to the threat of audio and visual forgeries. Most previous studies on detecting artificial intelligence-generated fake videos only utilize visual modality or audio modality. While some methods exploit audio and visual modalities to detect forged videos, they have not been comprehensively evaluated on multimodal datasets of deepfake videos involving acoustic and visual manipulations, and are mostly based on convolutional neural networks with low detection accuracy. Considering that human cognition instinctively integrates multisensory information including audio and visual cues to perceive and interpret content and the success of transformer in various fields, this study introduces the audio-visual transformer-based ensemble network (AVTENet). This innovative framework tackles the complexities of deepfake technology by integrating both acoustic and visual manipulations to enhance the accuracy of video forgery detection. Specifically, the proposed model integrates several purely transformer-based variants that capture video, audio, and audio-visual salient cues to reach a consensus in prediction. For evaluation, we use the recently released benchmark multimodal audio-video FakeAVCeleb dataset. For a detailed analysis, we evaluate AVTENet, its variants, and several existing methods on multiple test sets of the FakeAVCeleb dataset. Experimental results show that the proposed model outperforms all existing methods and achieves state-of-the-art performance on Testset-I and Testset-II of the FakeAVCeleb dataset. We also compare AVTENet against humans in detecting video forgery. The results show that AVTENet significantly outperforms humans. © 2016 IEEE.

关键词： audio-visual AVTENet Deepfake detection

来源：评论

学校读者我要写书评

暂无评论

Decomposition of dynamic transcriptomic responses during effector-triggered immunity reveals conserved responses in two distinct plant cell populations

引用

Plant Communications 2024年第8期5卷 1-17页

作者： Xiaotong Liu Daisuke Igarashi Rachel A.Hillmer Thomas Stoddard You Lu Kenichi Tsuda Chad L.Myers Fumiaki Katagiri Department of Plant and Microbial Biology University of Minnesota-Twin CitiesSt PaulMN 55108USA Department of Computer Science and Engineering University of Minnesota-Twin CitiesMinneapolisMN 55455USA Bioinformatics and Computational Biology Graduate Program University of Minnesota-Twin CitiesMinneapolisMN 55455USA Institute for Innovation Ajinomoto Co.Inc.KawasakiJapan State Key Laboratory of Agricultural Microbiology Hubei Hongshan LaboratoryHubei Key Lab of Plant PathologyCollege of Plant Science and TechnologyHuazhong Agricultural UniversityWuhan 430070China Present address:225 S 6th St MinneapolisMN 55402USA Present address:Biotalys Inc.2520 Meridian Parkway Suite 480DurhamNC 27713USA Present address:Life Sciences Institute and Department of Cell and Developmental Biology University of Michigan Medical SchoolAnn ArborMI 48109USA

Rapid plant immune responses in the appropriate cells are needed for effective defense against *** transcriptome analysis is often used to describe overall immune responses,collection of transcriptome data with sufficient resolution in both space and time is *** reanalyzed public Arabidopsis time-course transcriptome data obtained after low-dose inoculation with a Pseudomonas syringae strain expressing the effector AvrRpt2,which induces effector-triggered immunity in ***-peak time-course patterns are prevalent among thousands of upregulated *** implemented a multicompartment modeling approach to decompose the double-peak pattern into two single-peak patterns for each *** decomposed peaks reveal an“echoing”pattern:the peak times of the first and second peaks correlate well across most upregulated *** demonstrated that the two peaks likely represent responses of two distinct cell populations that respond either cell autonomously or indirectly to ***,the peak decomposition has extracted spatial information from the time-course *** echoing pattern also indicates a conserved transcriptome response with different initiation times between the two cell populations despite different elicitor types.A gene set highly overlapping with the conserved gene set is also upregulated with similar kinetics during pattern-triggered *** of a WRKY network via different entry-point WRKYs can explain the similar but not identical transcriptome responses elicited by different elicitor *** discuss potential benefits of the properties of the WRKY activation network as an immune signaling network in light of pressure from rapidly evolving pathogens.

关键词： Arabidopsis Pseudomonas syringae plant immune response multi-compartment model WRKY transcription factors resilient immune signaling network

来源：评论

学校读者我要写书评

暂无评论

eVision: Forecasting the spread of Tuberculosis in India with Deep Learning

eVision: Forecasting the spread of Tuberculosis in India wit...

引用

International Conference on Computational science and Computational Intelligence (CSCI)

作者： Juan Zuluaga Michael Castillo Divya Syal Andres Calle Navid Shaghaghi Department of Bioengineering (BIOE) Computer Science & Engineering (CSEN) Ethical Pragmatic & Intelligent Computing (EPIC) Laboratory in collaboration with the Healthcare Innovation & Design (HID) Program Information Systems & Analytics (ISA) and Mathematics & Computer Science (MCS) Santa Clara University Santa Clara California USA

Humanity has battled tuberculosis for all of recorded history. Some studies estimate that Mycobacterium tuberculosis may have been around as long as 3 million years but it was only in 1834 when Johann Schonlein officially presented the characteristics of it. Even though the TB epidemic has touched all corners of the world, Africa and Asia are the regions that currently suffer the worst consequences. The purpose of this study is to construct a model within the eVision forecasting environment, capable of forecasting the trend of Tuberculosis cases in India, as India is the country that accounts for the largest percentage of TB cases and deaths worldwide. And being able to make predictions for India may also lead to successful results for other regions in Asia and Africa. In order to do so, this study presents different test cases that show the effectiveness of the model, varying the number of steps for each one of the data sets created. It's important to note, that these data sets are combinations of data gathered from the states with the most TB cases in India in the last years, as well as the total data for India, and supplemental data from Google Trends, as a way to facilitate the machine learning model. Even though the final results were respectable compared to past research done on India and other countries, the model nevertheless has a limitation on the number of weeks ahead which the predictions are still considered to be good; with 7 weeks being the optimal result.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Preferences for Using Canva Apps as a Presentation Media Creation Tool Among Gen Z Students: SEM Analysis with TAM and UTAUT2 Approaches

Preferences for Using Canva Apps as a Presentation Media Cre...

引用

International Seminar on Intelligent Technology and Its Applications (ISITIA)

作者： Fairuz Iqbal Maulana Dian Lestari Agung Purnomo Rahman Arifuddin Vandha Pradwiyasma Widartha Puput Dani Prasetyo Adi Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Faculty of Medicine Airlangga Surabaya 60132 Indonesia Surabaya Indonesia Entrepreneurship Department BINUS Business School Undergraduate Program Bina Nusantara University Jakarta Indonesia Department of Electrical Engineering University of Merdeka Malang Malang Indonesia Information System of Industrial Engineering Faculty Telkom University Bandung Indonesia Research Center for Telecommunication National Research and Innovation Agency (BRIN) Bandung Indonesia

ISBN: (数字)9798350378573

ISBN: (纸本)9798350378580

The purpose of this study is to find out what makes Generation Z students accept and use Canva as a tool for making presentation materials. The conceptual framework of this study is the combination of "Technology Acceptance Model (TAM)" and the "Unified Theory of Acceptance and Use of Technology (UTAUT2)". "Structural Equation Modeling (SEM)" used for data analyzation. This research aims to understand the dynamics of Canva acceptance among students. The findings show that enjoyment, fun, and pleasure in using Canva, combined with its creative tools and design-centric approach, significantly resonate with Gen Z's preference for aesthetics. This research highlights that Canva is favored by students for creating engaging presentations.

关键词： Seminars Analytical models Technology acceptance model Accuracy Numerical analysis Media Mathematical models

来源：评论

学校读者我要写书评

暂无评论

Characteristics Fluctuation of Sub-3-nm Bulk FinFET Devices Induced by Random Interface Traps

Characteristics Fluctuation of Sub-3-nm Bulk FinFET Devices ...

引用

IEEE Conference on Nanotechnology

作者： Sekhar Reddy Kola Min-Hui Chuang Yiming Li Parallel and Scientific Computing Laboratory Electrical Engineering and Computer Science International Graduate Program National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Electronics and Electrical Engineering Institute of Communications Engineering Institute of Biomedical Engineering Institute of Pioneer Semiconductor Innovation Institute of Artificial Intelligence Innovation National Yang Ming Chiao Tung University Hsinchu Taiwan

Three-dimensional (3D) bulk fin-typed field effect transistors (FinFETs) have emerged as key devices that can scale down the technology node beyond 22-nm. However, the scaled devices have created new sources of fluctuation inherent in 3D geometry. The interface trap is one such fluctuation that is caused by the trapping and de-trapping of charge carriers and has an adverse effect on device characteristics and variability. In this work, we study impacts of random interface traps (RITs) on electrical characteristic of bulk FinFETs by using a 3D quantum-mechanically corrected device simulation. RIT position effects on short channel effects (SCEs) are examined with physical governed influence to show the major fluctuations. More than 50% reductions of the RITs-induced characteristic fluctuation of the germanium (Ge) devices are observed, compared with Si devices. The Ge ones can reduce SCE variations and exhibit high immunity to RITs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multivariate Time Series Anomaly Detection with Deep Learning Models Leveraging Inter-Variable Relationships

Multivariate Time Series Anomaly Detection with Deep Learnin...

引用

Cybersecurity Conference (SVCC), Silicon Valley

作者： Changmin Seong Dongjun Lim Jiho Jang Jonghoon Lee Jong-Geun Park Yun-Gyung Cheong Department of Computer Science Sungkyunkwan University Suwon South Korea Department of Artificial Intelligence Sungkyunkwan University Suwon South Korea Electronics and Telecommunications Research Institute Daejeon South Korea Department of AI Social Innovation Convergence Program Sungkyunkwan University Suwon South Korea

This paper presents a system for multivariate time series anomaly detection using deep learning, with an added module to reflect variable relationships. The system uses an autoencoder to extract latent variables that reflect the time series characteristics of the variables, and calculates variable importance using the similarities among the variables. To evaluate the proposed method, experiments were conducted using three similarity measures: cosine similarity, distance correlation, and DTW. Four time series datasets were used for evaluation, and the results showed that the proposed model outperformed the baseline model in HAI 22.04 and HAI 21.03 datasets. For the WADI dataset, the Fl-score improved only when using cosine similarity, while the TaPR-Fl score improved only when using DTW. However, no performance improvement was observed in the SWaT dataset. These results suggest that the effectiveness of utilizing inter-variable relationships is dependent on the characteristics of the data and the similarity calculation method employed. Therefore, a careful selection of the appropriate similarity calculation method for a given dataset is necessary to achieve optimal performance improvements.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient and secure semi-quantum private comparison protocol using three-particle GHZ-like states against participant attack

引用

Modern Physics Letters A 2024年第10期39卷

作者： Yang, Chun-Wei Huang, Yu-Yun Lin, Jason Tsai, Chia-Wei Master Program for Digital Health Innovation China Medical University No. 100 Section 1 Jingmao Road Beitun District Taichung 406040 Taiwan Department of Computer Science and Engineering National Chung Hsing University No. 145 Xingda Road South District Taichung 40227 Taiwan Department of Computer Science and Information Engineering National Taichung University of Science and Technology No. 129 Section 3 Sanmin Road North District Taichung 40401 Taiwan

This study identifies a participant attack vulnerability in Li et al.'s SQPC protocol. The participant attack allows a malicious participant, Bob, to obtain the participant Alice's secret information by intercepting and measuring photons sent to Alice, and later decrypting Alice's encrypted comparison result. An improved SQPC protocol is proposed to address this problem. The key enhancement is introducing a mutual eavesdropping check requiring the third party and participants to verify they obtain the same photon measurement results. This enables detecting participant attacks like the one identified. The SQPC protocol proposed in this study not only withstands attacks from internal participants but also achieves quantum efficiency almost equivalent to Li et al.'s SQPC protocol. Thus, this study presents a more secure SQPC protocol without compromising quantum efficiency. © 2024 World Scientific Publishing Company.

关键词： GHZ-like state participant attack Quantum cryptography semi-quantum private comparison

来源：评论

学校读者我要写书评

暂无评论

Twitter Data Sentiment Analysis of COVID-19 Vaccination using Machine Learning

Twitter Data Sentiment Analysis of COVID-19 Vaccination usin...

引用

International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)

作者： Fairuz Iqbal Maulana Puput Dani Prasetyo Adi Dian Lestari Agung Purnomo Sukeipah Yuli Prihatin Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia National Research and Innovation Agency Telecommunication Research Center Jakarta Indonesia Faculty of Medicine Airlangga University Surabaya Indonesia Entrepreneurship Department BINUS Business School Undergraduate Program Bina Nusantara University Jakarta Indonesia Informatics Study Program Politeknik Kota Malang Malang Indonesia

The expansion of the web is accelerating, which helps encourage the creation of fresh ideas. In today's internet era, we must suggest techniques to filter out various information. Social media sentiment analysis based on Twitter data can monitor the real-time monitoring of the COVID-19 vaccine. In this way, relevant organizations or governments can take proactive steps to address misinformation and inappropriate behaviour around the COVID-19 vaccine, which threatens the success of the national vaccination campaign. The purpose of this research is to determine if there is a link between how people feel about the COVID-19 vaccine on Twitter and how many people actually get vaccinated against it. This study uses the COVID-19 All Vaccines Tweet dataset sourced from Kaggle. This research Identifies public sentiment, emotion, word usage, and trend of all filtered tweets. The results show that there are 31% positive tweets, 10% negative tweets, and 58% neutral tweets. Tweets with neutral subjective valence tend to cluster in the middle of the polarity scale (between -1 and +1), whereas tweets with strong subjective valence are spread across the scale.

关键词： COVID-19 Seminars Sentiment analysis Social networking (online) Blogs Tag clouds Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Unmasking Illusions: Understanding Human Perception of Audiovisual Deepfakes

arXiv

引用

arXiv 2024年

作者： Hashmi, Ammarah Shahzad, Sahibzada Adil Lin, Chia-Wen Tsao, Yu Wang, Hsin-Min Social Networks and Human-Centered Computing Program Taiwan International Graduate Program Institute of Information Science Academia Sinica Taipei11529 Taiwan Institute of Information Systems and Applications National Tsing Hua University Hsinchu300044 Taiwan Department of Computer Science National Chengchi University Taipei11529 Taiwan Department of Electrical Engineering Institute of Communications Engineering National Tsing Hua University Hsinchu300044 Taiwan Research Center for Information Technology Innovation Academia Sinica Taipei11529 Taiwan Institute of Information Science Academia Sinica Taipei11529 Taiwan

The emergence of contemporary deepfakes has attracted significant attention in machine learning research, as artificial intelligence (AI) generated synthetic media increases the incidence of misinterpretation and is difficult to distinguish from genuine content. Currently, machine learning techniques have been extensively studied for automatically detecting deepfakes. However, human perception has been less explored. Malicious deepfakes could ultimately cause public and social problems. Can we humans correctly perceive the authenticity of the content of the videos we watch? The answer is obviously uncertain;therefore, this paper aims to evaluate the human ability to discern deepfake videos through a subjective study. We present our findings by comparing human observers to five state-of-the-art deepfake detection models. To this end, we used gamification concepts to provide 110 participants (55 native English speakers and 55 non-native English speakers) with a web-based platform where they could access a series of 40 videos (20 real and 20 fake) to determine their authenticity. Each participant performed the experiment twice with the same 40 videos in different random orders. The videos are manually selected from the FakeAVCeleb dataset. We found that all AI models performed significantly better than humans when evaluated on the same 40 videos. The study also reveals that while deception is not impossible, humans tend to overestimate their detection capabilities. Our experimental results may help benchmark human versus machine performance, advance forensics analysis, and enable adaptive countermeasures. © 2024, CC BY.

关键词： Audiovisual

来源：评论

学校读者我要写书评

暂无评论

Understanding Audiovisual Deepfake Detection: Techniques, Challenges, Human Factors and Perceptual Insights

arXiv

引用

arXiv 2024年

作者： Hashmi, Ammarah Shahzad, Sahibzada Adil Lin, Chia-Wen Tsao, Yu Wang, Hsin-Min The Social Networks and Human-Centered Computing Program Taiwan International Graduate Program Institute of Information Science Academia Sinica Taipei11529 Taiwan The Institute of Information Systems and Applications National Tsing Hua University Hsinchu300044 Taiwan The Department of Computer Science National Chengchi University Taipei11529 Taiwan The Department of Electrical Engineering The Institute of Communications Engineering National Tsing Hua University Hsinchu300044 Taiwan The Research Center for Information Technology Innovation Academia Sinica Taipei11529 Taiwan The Institute of Information Science Academia Sinica Taipei11529 Taiwan

Deep Learning has been successfully applied in diverse fields, and its impact on deepfake detection is no exception. Deepfakes are fake yet realistic synthetic content that can be used deceitfully for political impersonation, phishing, slandering, or spreading misinformation. Despite extensive research on unimodal deepfake detection, identifying complex deepfakes through joint analysis of audio and visual streams remains relatively unexplored. To fill this gap, this survey first provides an overview of audiovisual deepfake generation techniques, applications, and their consequences, and then provides a comprehensive review of state-of-the-art methods that combine audio and visual modalities to enhance detection accuracy, summarizing and critically analyzing their strengths and limitations. Furthermore, we discuss existing open source datasets for a deeper understanding, which can contribute to the research community and provide necessary information to beginners who want to analyze deep learning-based audiovisual methods for video forensics. By bridging the gap between unimodal and multimodal approaches, this paper aims to improve the effectiveness of deepfake detection strategies and guide future research in cybersecurity and media integrity. © 2024, CC BY.

关键词： Audiovisual

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：