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

Deep and shallow data science for multi-scale optical neuroscience

arXiv

引用

arXiv 2024年

作者： Mishne, Gal Charles, Adam Halıcıoğlu Data Science Institute Department of Electrical and Computer Engineering the Neurosciences Graduate Program UC San Diego 9500 Gilman Drive La Jolla CA92093 United States Department of Biomedical Engineering Kavli Neuroscience Discovery Institute Center for Imaging Science Department of Neuroscience Mathematical Institute for Data Science Johns Hopkins University BaltimoreMD21287 United States

Optical imaging of the brain has expanded dramatically in the past two decades. New optics, indicators, and experimental paradigms are now enabling in-vivo imaging from the synaptic to the cortex-wide scales. To match the resulting flood of data across scales, computational methods are continuously being developed to meet the need of extracting biologically relevant information. In this pursuit, challenges arise in some domains (e.g., SNR and resolution limits in micron-scale data) that require specialized algorithms. These algorithms can, for example, make use of state-of-the-art machine learning to maximally learn the details of a given scale to optimize the processing pipeline. In contrast, other methods, however, such as graph signal processing, seek to abstract away from some of the details that are scale-specific to provide solutions to specific sub-problems common across scales of neuroimaging. Here we discuss limitations and tradeoffs in algorithmic design with the goal of identifying how data quality and variability can hamper algorithm use and dissemination. © 2024, CC BY.

关键词： Neuroimaging

来源：评论

学校读者我要写书评

暂无评论

A Detailed Analysis of Qualitative and Quantitative Factors in Realization of 6G Communication

A Detailed Analysis of Qualitative and Quantitative Factors ...

引用

International Conference on Internet of Things: Systems, Management and Security

作者： Mohammad Alja’Afreh Ali Karime Sahel Alouneh Abdulmotaleb El Saddik Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI) UAE School of Electrical Engineering and Computer Science University of Ottawa Canada Department of Electrical and Computer Engineering The Royal Military College of Canada Canada Cybersecurity Program College of Engineering AL AIN University UAE Computer Engineering Department German Jordanian University

Undoubtedly, the world has so far faced a pandemic which is reshaping daily lives and business activities. Even at current endemic stages, special focus on maintaining physical distancing norms for curbing the expeditious spread of the disease, many institutions, individuals and industries rely on communications networks or telecoms for ensuring service consistency to avoid complete termination of their business operations and other activities. This has put enormous pressure on mobile networks and communication systems thereby making the technology experts to think more about introducing rapid speed, vast coverage and high connectivity networks. The extensive application of fresh communication networks and enabling technologies have impelled the advent of 6G communication networks. As 6G is still in its inception phase, its complete realization requires a proper and high understanding of diverse quantitative and qualitative factors supporting its deployment. From this standpoint, this survey article intends to deeply explore 6G networks, their significance and prerequisites. This paper provides a succinct theoretical background of 6G technology, and reviews the diverse enabling technologies and existing works undertaken on core technologies. It explores the prevailing gaps in research for providing readers to gain information regarding challenges in perfect 6G network realization and implementation thus paving the road for a successful 6G vision.

关键词： 6G mobile communication Systematics Pandemics Roads Telecommunications Communication networks Security

来源：评论

学校读者我要写书评

暂无评论

Inkwell: Design and Validation of a Low-Cost Open Electricity-Free 3D Printed Device for Automated Thin Smearing of Whole Blood

arXiv

引用

arXiv 2023年

作者： Nowak, Jerome Kothari, Anesta Li, Hongquan Pannu, Jaspreet Algazi, Dani Prakash, Manu Department of Mechanical Engineering United States Department of Bioengineering United States Department of Electrical Engineering United States Department of Medicine United States Department of Computer Science United States United States Woods Institute of the Environment United States Center for Innovation in Global Health United States Stanford University StanfordCA United States

Microscopy plays a crucial role in hematology and diagnosis of infectious diseases worldwide. For malaria alone, more than 200 million slides are read by manual microscopists every year. High quality thin blood smears are essential for subsequent microscopy examinations including malaria microscopy, but are hard to make in field settings. Current manual smearing methods lack consistency and often do not provide a uniformly dense mono-layer of red blood cells, even when prepared by trained experts. Existing devices for assisting in making thin smears are available but are limited by cost or complexity for wider use. Here we present Inkwell, a portable mechanical device capable of making high quality thin blood smears in field settings. Inkwell is simple, low-cost, does not use electricity, and requires minimal training prior to use. By utilizing passive dissipative dynamics of a spiral spring coupled to an air dashpot with a tunable valve - we demonstrate a highly tunable mechanism for constant velocity smears at prescribed angle. Inkwell is capable of producing high quality blood smears of tunable cell density with more than 12 million individually distinguishable red blood cells on a single slide. The current design, which exploits precision manufacturing of a 17 cents plastic syringe and a spring, can be printed on a standard 3D printer with overall unit cost of less than a few dollars in large quantities. We further present usability tests to confirm performance over 10,000 unit cycle operations with no degradation in quality of the smear and demonstrate ease of use with minimal training. Inkwell enhances the broader toolbox of open innovations in diagnostics for providing high quality medical care in low and medium resource settings. Combined with rise of 3D printing, Inkwell presents an alternative to traditional centralized manufacturing and opens up distributed manufacturing of medical diagnostics in global context. © 2023, CC BY.

关键词： 3D printing

来源：评论

学校读者我要写书评

暂无评论

Study of ASIC Readout Used for Detectors in an MR Compatible, RF-Penetrable TOF-PET Insert

Study of ASIC Readout Used for Detectors in an MR Compatible...

引用

IEEE Symposium on Nuclear science (NSS/MIC)

作者： Qian Dong Salar Sajedi Ilaria Sacco Peter Fischer Craig S. Levin Departments of Electrical Engineering and Radiology Stanford University Stanford CA USA Departments of Radiology Stanford University Stanford CA USA Department of Computer Science Heidelberg University Baden-Württemberg Germany Department of Radiology Electrical Engineering Bioengineering and Physics Stanford University Molecular Imaging Program at Stanford (MIPS) Stanford CA USA

Developing positron emission tomography (PET) systems with high channel number has driven the need for application-specific integrated circuits (ASICs) owing to their compactness, multi-channel readout capability, and excellent time and energy resolution performance. They can also be quite flexible by employing field-programmable gate arrays (FPGAs) in the readout circuit. In particular, the ASIC+FPGA approach provides a high performance and compact solution for TOF-PET applications. ASICs serve functions of extracting time and energy information from analog SiPM signals, and an FPGA is employed for other application-specific requirements like on-board chip configuration, communication with data acquisition (DAQ) systems, and system monitoring. Based on this foundation, we developed TOF-capable detector modules for our second generation RF-penetrable PET insert. During our previous experiments, we found that different trigger and readout methods lead to different noise pattern and intensity in MR environment, and also have different singles count loss ratio. This paper aims to study and compare different readout configurations for our MR-compatible detectors. We studied four trigger and readout methods: hit-standard, frequent-standard, hit-bypass, and frequent-bypass, and compared their singles count loss ratios with both theoretical models and experimental results. For all four methods, the theoretical results matched the experimental results well. The two methods which used bypass mode performed better than those using the standard mode approach. With the most optimized trigger and readout method (hit-bypass), we achieved 1.2% singles count loss ratio at a 20 kcps count rate per channel.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Resource Allocation in Software-Defined Networks for IoT Application

A Novel Resource Allocation in Software-Defined Networks for...

引用

International Conference on Software, Telecommunications and computer Networks (SoftCOM)

作者： Alexandre Ladeira de Sousa Ogobuchi Daniel OKey Renata Lopes Rosa Muhammad Saadi Demostenes Zegarra Rodriguez Department of Computer Science Federal University of Lavras Lavras Minas Gerais Brazil Graduate Program in Information Engineering Federal University of ABC Santo Andre Sao Paulo Brazil Department of Electrical Enginnering University of Central Punjab Punjab Pakistan Computer Science Department Federal University of Lavras Lavras Minas Gerais Brazil

The rapid proliferation of Internet of Things (IoT) devices has posed significant challenges for network resource allocation and management. In this paper, we propose a novel methodology for efficient resource allocation in Software-Defined Networks (SDNs) to meet the unique requirements of IoT applications. Our approach leverages the flexibility and programmability of SDN to dynamically allocate network resources based on the specific needs of IoT devices. Our proposal combines reinforcement learning with dynamic resource allocation, demonstrating its effectiveness in optimizing resource allocation and enhancing network performance. The results highlight the potential of our approach to addressing the resource challenges in SDN-based IoT environments. The proposed method obtains results superior to 30% of throughput in comparison to a traditional resource allocation in SDN to IoT applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unified Approach to Video-Based AI Inference Tasks in Augmented Reality Systems Assisted by Mobile Edge Computing

Unified Approach to Video-Based AI Inference Tasks in Augmen...

引用

International Conference on Software, Telecommunications and computer Networks (SoftCOM)

In the rapidly evolving field of Augmented Reality (AR), delivering real-time, immersive experiences places a significant demand on computational resources, particularly in the context of video-based Artificial Intelligence (AI) inference tasks. Some solutions explorate optimize the tasks using the potential of Mobile Edge Computing (MEC), in which MEC brings the advantages of reduced latency and network congestion, which are critical for AR systems. In this work, we designed a novel algorithm named AR-MEC Optimization for the distribution of AI inference tasks among edge servers while considering several factors including task complexity, server capacity, and network conditions. Our approach not only ensures low-latency responses but also improves overall system efficiency and the user’s Quality of Experience (QoE). Experimental results demonstrate that our unified approach significantly outperforms traditional cloud-based methods in terms of response time and computational efficiency, thereby highlighting the potential of MEC-assisted AR systems for real-world applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

BRAIN-TO-TEXT BENCHMARK’24: LESSONS LEARNED

arXiv

引用

arXiv 2024年

作者： Willett, Francis R. Li, Jingyuan Le, Trung Fan, Chaofei Chen, Mingfei Shlizerman, Eli Chen, Yue Zheng, Xin Okubo, Tatsuo S. Benster, Tyler Lee, Hyun Dong Kounga, Maxwell Buchanan, E. Kelly Zoltowski, David Linderman, Scott W. Henderson, Jaimie M. Department of Neurosurgery Stanford University United States Department of Electrical & Computer Engineering University of Washington United States Department of Computer Science Stanford University United States Department of Applied Mathematics University of Washington United States China Neurosciences Phd Program Stanford University United States Wu Tsai Neurosciences Institute Department of Statistics Stanford University United States

Speech brain-computer interfaces aim to decipher what a person is trying to say from neural activity alone, restoring communication to people with paralysis who have lost the ability to speak intelligibly. The Brain-to-Text Benchmark’24 and associated competition was created to foster the advancement of decoding algorithms that convert neural activity to text. Here, we summarize the lessons learned from the competition ending on June 1, 2024 (the top 4 entrants also presented their experiences in a recorded webinar). The largest improvements in accuracy were achieved using an ensembling approach, where the output of multiple independent decoders was merged using a fine-tuned large language model (an approach used by all 3 top entrants). Performance gains were also found by improving how the baseline recurrent neural network (RNN) model was trained, including by optimizing learning rate scheduling and by using a diphone training objective. Improving upon the model architecture itself proved more difficult, however, with attempts to use deep state space models or transformers not yet appearing to offer a benefit over the RNN baseline. The benchmark will remain open indefinitely to support further work towards increasing the accuracy of brain-to-text algorithms. © 2024, CC BY.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Unbiased Approximations for Stationary Distributions of McKean-Vlasov SDEs

arXiv

引用

arXiv 2024年

作者： Awadelkarim, Elsiddig Chada, Neil K. Jasra, Ajay Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia Department of Mathematics City University of Hong Kong Hong Kong School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen China

We consider the development of unbiased estimators, to approximate the stationary distribution of Mckean-Vlasov stochastic differential equations (MVSDEs). These are an important class of processes, which frequently appear in applications such as mathematical finance, biology and opinion dynamics. Typically the stationary distribution is unknown and indeed one cannot simulate such processes exactly. As a result one commonly requires a time-discretization scheme which results in a discretization bias and a bias from not being able to simulate the associated stationary distribution. To overcome this bias, we present a new unbiased estimator taking motivation from the literature on unbiased Monte Carlo. We prove the unbiasedness of our estimator, under assumptions. In order to prove this we require developing ergodicity results of various discrete time processes, through an appropriate discretization scheme, towards the invariant measure. Numerous numerical experiments are provided, on a range of MVSDEs, to demonstrate the effectiveness of our unbiased estimator. Such examples include the Currie-Weiss model, a 3D neuroscience model and a parameter estimation problem. © 2024, CC BY-NC-SA.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Optoelectric Raman Nanosensors: Overcoming Intrinsic Limit in Nano-biosensing

Research Square

引用

Research Square 2024年

作者： Lian, Bin Li, Huaizhi Guo, Jianhe Fan, Donglei Emma Materials Science and Engineering Program Texas Materials Institute The University of Texas AustinTX78712 United States Walker Department of Mechanical Engineering Texas Materials Institute The University of Texas at Austin AustinTX78712 United States Chandra Family Department of Electrical and Computer Engineering The University of Texas at Austin AustinTX78712 United States

It is intrinsically difficult to detect low-concentration biomolecules with both ultra-sensitivity and high speed for early-stage disease diagnosis. The challenge originates from the small size of nanosensors, which enables ultra-sensitivity biosensing, while also substantially increases the detection time of dilute molecules. in this work, we report an original optoelectric sensing scheme, along with an innovative device design and a strategic fabrication approach to overcome such a challenge in biosensing. The proposed scheme exploits the profound optoelectric effect of semiconductor silicon nanowires, which can readily enrich trace-amounts of biomolecules at the point of laser and simultaneously detect Raman signals of focused molecules. The device design includes large arrays of silicon nanorods with electrodes integrated at their roots, and surface-distributed dense plasmonic silver nanoparticles for surface-enhanced Raman spectroscopy (SERS) detection. Operating at only -0.8 V in an electrochemical cell, these optoelectric nanosensors readily achieve a 150-fold signal enhancement, improving the detection limit of probing molecules, adenine, by five orders of magnitude, to 0.6 fM. The enhancement effect is robust, observed across concentrations from 1 μM to 1 fM. The working mechanism is general, not only for detecting small molecules, such as adenine, but also for large charged molecules, such as Vertebrate DNA. The underlying novel mechanism is unraveled by multiscale numerical simulations and calculations. This research, addressing an arduous bottleneck issues in nanobiosensing, is expected to inspire a new class of biochemical sensors, important for the practical detection of trace-amount biochemicals in solution, important for clinic relevant early-stage disease diagnostics. © 2024, CC BY.

关键词： Nanorods

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：