检索结果-内蒙古大学图书馆

Deep Fake Detection using Transfer Learning: A Comparative study of Multiple Neural Networks

Deep Fake Detection using Transfer Learning: A Comparative s...

2024 International Conference on Signal Processing, Computation, Electronics, Power and Telecommunication, IConSCEPT 2024

作者： Abhineswari, M. Charan, Kurupati Sai Bn, Shrikarti Kanmani R, Sujithra Shool of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798331540685

With the proliferation of sophisticated AI techniques, the creation and dissemination of deep fake images and videos have emerged as a pressing concern in today's digital landscape. Deep fake technology employs advanced machine learning algorithms to manipulate or synthesize realistic-looking media, often with malicious intent. This study evaluates the effectiveness of various pre-trained deep learning models using transfer learning for detecting deep fake images on the Face Forensics++ dataset. The models considered include MobileN etV2, ResN et50, Inception V3, EfficientN et, Xception, NASNetMobile, and a Custom CNN. Accuracies obtained from these models are compared to assess their performance in distinguishing between real and fake images. With the highest performance in MobileNetV2 with 89% followed by ResNet50 with 83 % and other models. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Embodied Contrastive Learning with Geometric Consistency and Behavioral Awareness for Object Navigation 24

Embodied Contrastive Learning with Geometric Consistency and...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Chen, Bolei Kang, Jiaxu Zhong, Ping Liang, Yixiong Sheng, Yu Wang, Jianxin Shool of Computer Science and Engineering Central South University Changsha China

ISBN: (纸本)9798400706868

Object Navigation (ObjcetNav), which enables an agent to seek any instance of an object category specified by a semantic label, has shown great advances. However, current agents are built upon occlusion-prone visual observations or compressed 2D semantic maps, which hinder their embodied perception of 3D scene geometry and easily lead to ambiguous object localization and blind exploration. To address these limitations, we present an Embodied Contrastive Learning (ECL) method with Geometric Consistency (GC) and Behavioral Awareness (BA), which motivates agents to actively encode 3D scene layouts and semantic cues. Driven by our embodied exploration strategy, BA is modeled by predicting navigational actions based on multi-frame visual images, as behaviors that cause differences between adjacent visual sensations are crucial for learning correlations among continuous visions. The GC is modeled as the alignment of behavior-aware visual stimulus with 3D semantic shapes by employing unsupervised contrastive learning. The aligned behavior-aware visual features and geometric invariance priors are injected into a modular ObjectNav framework to enhance object recognition and exploration capabilities. As expected, our ECL method performs well on object detection and instance segmentation tasks. Our ObjectNav strategy outperforms state-of-the-art methods on MP3D and Gibson datasets, showing the potential of our ECL in embodied navigation. © 2024 ACM.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Efficient-Learning Grasping and Pushing in Dense Stacking via Mask Function and Pixel Overlap Rate 7

Efficient-Learning Grasping and Pushing in Dense Stacking vi...

引用

7th International Symposium on Autonomous Systems, ISAS 2024

作者： Lian, Jie Jiang, Juncheng Qiu, Chaochao Pan, Qinghui Dong, Yongxiang Wang, Zhao Wang, Dong Dalian University of Technology School of Control Science and Engineering Dalian China University of Science and Technology Liaoning Shool of Computer Science and Software Engineering China

ISBN: (纸本)9798350363173

In the field of robotic grasping, grasping tightly stacked objects is a formidable challenge. The presence of non-target objects significantly increases the risk of grasping failure. To solve this problem, the aim is to take action while grasping, reduce clutter, and effectively complete grasping tasks. Based on deep reinforcement learning, the Efficient Learning Grasping(ELG) framework is proposed that integrates pushing actions alongside grasping to efficiently pick up objects from cluttered environments with reduced training time. However, in the process of grasping, the agent is rewarded only when the grasp is successful, so the data collection is extremely inefficient and spends a lot of time on the training of grasping strategy. In our work, two strategies are proposed to improve the training efficiency. First, two mask functions are introduced as prior knowledge to enable the agent to focus on more meaningful regions while ignoring some unnecessary regions, aiming at effective learning. Second, The Pixel Overlap Rate (POR) is introduced to quantify the environment clutter level, and set a new push action reward function based on the POR. The implementation of the above ideas in a simulation environment and transferred the learned models to the real world and verified them in practice. Compared with existing method, our method improves the training efficiency of the initial algorithm while having a higher success rate of grasping. Detailed grasping video at https://***/CqtSvaVl60M. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

VisualRWKV: Exploring Recurrent Neural Networks for Visual Language Models 31

VisualRWKV: Exploring Recurrent Neural Networks for Visual L...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Hou, Haowen Zeng, Peigen Ma, Fei Yu, Fei Richard Shenzhen China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Shool of Information Technology Carleton University Canada

ISBN: (纸本)9798891761964

Visual Language Models (VLMs) have rapidly progressed with the recent success of large language models. However, there have been few attempts to incorporate efficient linear Recurrent Neural Networks (RNNs) architectures into VLMs. In this study, we introduce VisualRWKV, the first application of a linear RNN model to multimodal learning tasks, leveraging the pre-trained RWKV language model. We propose a data-dependent recurrence and sandwich prompts to enhance our modeling capabilities, along with a 2D image scanning mechanism to enrich the processing of visual sequences. Extensive experiments demonstrate that VisualRWKV achieves competitive performance compared to Transformer-based models like LLaVA-1.5 on various benchmarks. Compared to LLaVA-1.5, VisualRWKV has a speed advantage of 3.98 times and can save 54% of GPU memory when reaching an inference length of 24K tokens. To facilitate further research and analysis, we have made the checkpoints and the associated code publicly accessible at the following GitHub repository: https://***/howard-hou/VisualRWKV. © 2025 Association for Computational Linguistics.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Personalized Privacy-Preserving Information Fusion in Mobile Crowdsensing 20

Personalized Privacy-Preserving Information Fusion in Mobile...

引用

20th IEEE International Symposium on Parallel and Distributed Processing with Applications, 12th IEEE International Conference on Big Data and Cloud Computing, 12th IEEE International Conference on Sustainable Computing and Communications and 15th IEEE International Conference on Social Computing and Networking, ISPA/BDCloud/SocialCom/SustainCom 2022

作者： Feng, Zheqi Peng, Tao Wang, Guojun Guan, Kejian Guangzhou University Shool of Computer Science and Cyber Engineering Guangzhou510006 China

ISBN: (纸本)9781665464970

In mobile crowdsensing (MCS), information fusion is necessary to obtain data aggregation results. In information fusion, most of the existing personalized mechanisms only consider meeting the personalized requirements of data collectors and rarely consider the potential personalized requirements of data users. Furthermore, how to provide personalized data aggregation results for data users with different data quality requirements has not been addressed. This paper proposes a personalized privacy-preserving information fusion (P3IF) mechanism based on the personalized sampling mechanism in the MCS. P3IF provides personalized data aggregation results for data users with different data quality requirements while ensuring personalized privacy protection for data collectors. Specifically, we design a personalized privacy-preserving information fusion scheme based on the personalized differential privacy mechanism. Considering on the privacy requirements of the data collectors, we conduct personalized privacy sampling on the raw data from the data collectors and obtain a subset of sensing data. According to the data quality requirements of data users and the quality of each data, the obtained subset is sampled for personalized data quality requirements, and the final data set is obtained. Simulation experiments verify the feasibility and effectiveness of P3IF. © 2022 IEEE.

关键词： Privacy-preserving techniques

来源：评论

学校读者我要写书评

暂无评论

Deep Fake Detection using Transfer Learning: A Comparative study of Multiple Neural Networks

Deep Fake Detection using Transfer Learning: A Comparative s...

引用

Signal Processing, Computation, Electronics, Power and Telecommunication (IConSCEPT), International Conference on

作者： Abhineswari M Kurupati Sai Charan Shrikarti BN Sujithra Kanmani R Shool of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (数字)9798331540685

ISBN: (纸本)9798331540692

关键词： Deepfakes Accuracy Computational modeling Transfer learning Neural networks Pressing Signal processing

来源：评论

学校读者我要写书评

暂无评论

Efficient-Learning Grasping and Pushing in Dense Stacking via Mask Function and Pixel Overlap Rate

Efficient-Learning Grasping and Pushing in Dense Stacking vi...

引用

International Symposium on Autonomous Systems (ISAS)

作者： Jie Lian Juncheng Jiang Chaochao Qiu Qinghui Pan Yongxiang Dong Zhao Wang Dong Wang School of Control Science and Engineering Dalian University of Technology Dalian China Shool of Computer Science and Software Engineering University of Science and Technology Liaoning

ISBN: (数字)9798350363173

ISBN: (纸本)9798350363180

关键词： Training Stacking Grasping Production Data collection Deep reinforcement learning Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

VisualRWKV: Exploring Recurrent Neural Networks for Visual Language Models

arXiv

引用

arXiv 2024年

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Personalized Privacy-Preserving Information Fusion in Mobile Crowdsensing

Personalized Privacy-Preserving Information Fusion in Mobile...

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Zheqi Feng Tao Peng Guojun Wang Kejian Guan Shool of Computer Science and Cyber Engineering Guangzhou University Guangzhou China

ISBN: (数字)9781665464970

ISBN: (纸本)9781665464987

关键词： Differential privacy Crowdsensing Data integrity Data aggregation Sensors Task analysis

来源：评论

学校读者我要写书评

暂无评论

From Handcrafted Features to LLMs: A Brief Survey for Machine Translation Quality Estimation

From Handcrafted Features to LLMs: A Brief Survey for Machin...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Haofei Zhao Yilun Liu Shimin Tao Weibin Meng Yimeng Chen Xiang Geng Chang Su Min Zhang Hao Yang Shool of Computer Science and Engineering Northeastern University Shenyang China Huawei China National Key Laboratory for Novel Software Technology Nanjing University Nanjing China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Machine Translation Quality Estimation (MTQE) is the task of estimating the quality of machine-translated text in real time without the need for reference translations, which is of great importance for the development of MT. After two decades of evolution, QE has yielded a wealth of results. This article provides a comprehensive overview of QE datasets, annotation methods, shared tasks, methodologies, challenges, and future research directions. It begins with an introduction to the background and significance of QE, followed by an explanation of the concepts and evaluation metrics for word-level QE, sentence-level QE, document-level QE, and explainable QE. The paper categorizes the methods developed throughout the history of QE into those based on handcrafted features, deep learning, and Large Language Models (LLMs), with a further division of deep learning-based methods into classic deep learning and those incorporating pre-trained language models (LMs). Additionally, the article details the advantages and limitations of each method and offers a straightforward comparison of different approaches. Finally, the paper discusses the current challenges in QE research and provides an outlook on future research directions.

关键词： Deep learning Learning systems Surveys Measurement Annotations Neural networks Estimation

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：