检索结果-内蒙古大学图书馆

MFCLIP: Multi-modal Fine-grained CLIP for Generalizable Diffusion Face Forgery Detection

IEEE Transactions on Information Forensics and Security 2025年 20卷 5888-5903页

作者： Zhang, Yaning Wang, Tianyi Yu, Zitong Gao, Zan Shen, Linlin Chen, Shengyong Qilu University of Technology Shandong Academy of Sciences Faculty of Computer Science and Technology Jinan 250014 China National University of Singapore School of Computing 21 Lower Kent Ridge Rd 119077 Singapore Great Bay University School of Computing and Information Technology Dongguan 523000 China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen 518060 China Qilu University of Technology (Shandong Academy of Sciences) Shandong Artificial Intelligence Institute Jinan 250014 China Tianjin University of Technology Key Laboratory of Computer Vision and System Ministry of Education Tianjin 300384 China Shenzhen University Computer Vision Institute College of Computer Science and Software Engineering Shenzhen 518060 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen 518129 China Shenzhen University Guangdong Key Laboratory of Intelligent Information Processing Shenzhen 518060 China

The rapid development of photo-realistic face generation methods has raised significant concerns in society and academia, highlighting the urgent need for robust and generalizable face forgery detection (FFD) techniques. Although existing approaches mainly capture face forgery patterns using image modality, other modalities like fine-grained noises and texts are not fully explored, which limits the generalization capability of the model. In addition, most FFD methods tend to identify facial images generated by GAN, but struggle to detect unseen diffusion-synthesized ones. To address the limitations, we aim to leverage the cutting-edge foundation model, contrastive language-image pre-training (CLIP), to achieve generalizable diffusion face forgery detection (DFFD). In this paper, we propose a novel multi-modal fine-grained CLIP (MFCLIP) model, which mines comprehensive and fine-grained forgery traces across image-noise modalities via language-guided face forgery representation learning, to facilitate the advancement of DFFD. Specifically, we devise a fine-grained language encoder (FLE) that extracts fine global language features from hierarchical text prompts. We design a multi-modal vision encoder (MVE) to capture global image forgery embeddings as well as fine-grained noise forgery patterns extracted from the richest patch, and integrate them to mine general visual forgery traces. Moreover, we build an innovative plug-and-play sample pair attention (SPA) method to emphasize relevant negative pairs and suppress irrelevant ones, allowing cross-modality sample pairs to conduct more flexible alignment. Extensive experiments and visualizations show that our model outperforms the state of the arts on different settings like cross-generator, cross-forgery, and cross-dataset evaluations. © 2005-2012 IEEE.

关键词： CLIP Diffusion face forgery detection Image-noise fusion Sample pair attention Transformer

来源：评论

学校读者我要写书评

暂无评论

A Diffusion Probabilistic Model Based on Multi-Residualattention For Medical Image Segmentation

A Diffusion Probabilistic Model Based on Multi-Residualatten...

引用

International Conference on Wavelet Active Media Technology and Information Processing (ICWAMTIP)

作者： Yang Peng Li Chaorong Ling Xudong Qin Fengqing Zheng Yong Qiu Lihua School of Computer Science and Engineering Chongqing University of Technology Banan China Department of Artificial Intelligence and Big Data Yibin College YiBin China Chengdu Medical University Cheng Du China Yibin City second People's Hospital YiBin China

Diffusion models are currently among the most attention- grabbing generative models, achieving revolutionary applications across multiple domains and demonstrating exceptional performance. In the application of conditional diffusion models to alter specific regions of the output image to match desired features, most methods employ V AE encoders for feature extraction from conditional images. However, V AE encoders face challenges of difficult training and large parameter sizes; simultaneously, the denoising UNet in diffusion models often leads to blurred edges in sampled images. To effectively address these issues, We have introduced ResAtMedDiff, an innovative diffusion model specifically designed for the field of medical image processing, offering a novel approach to enhance the accuracy and efficiency of image analysis. The ResAtMedDiff model consists of two core training networks: the MEUNet conditional encoder and the RsAtUNet encoder-decoder. MEUNet is responsible for efficiently encoding the conditional images and then fusing these encodings with noise image features across multiple scales. This fusion strategy significantly enhances the ability of conditional features to guide the generation of noise images. The MEUN et encoder blends the features of the original conditional image with the Gaussian noise image features of the corresponding layers of the RsAtUN et encoder, followed by the RsAtUNet decoder decoding these fused features. This process achieves the generation of image segmentation masks guided by the features of the original image as medical prior conditions, As a result, this significantly enhances the precision of lesion area segmentation in medical imaging. The experimental findings reveal that the ResAtMedDiff model surpasses existing state-of-the-art segmentation techniques in its performance, particularly in the segmentation of thyroid nodules in ultrasound images and brain tumors in MRI scans, demonstrating its superior efficacy and precis

关键词：

来源：评论

学校读者我要写书评

暂无评论

Addressing the traveling salesperson problem with frequency fitness assignment and hybrid algorithms

引用

Soft Computing 2024年第17-18期28卷 9495-9508页

作者： Liang, Tianyu Wu, Zhize Lässig, Jörg Berg, Daan van den Thomson, Sarah L. Weise, Thomas Institute of Applied Optimization School of Artificial Intelligence and Big Data Hefei University Jinxiu Dadao 99 Anhui Hefei230601 China Fakultät Elektrotechnik und Informatik Hochschule Zittau/Görlitz Brückenstraße 1 Saxony Görlitz02826 Germany Abteilung Kognitive Energiesysteme Fraunhofer IOSB-AST Wilhelmsplatz 11 Saxony Görlitz02826 Germany Department of Computer Science Vrije Universiteit Amsterdam De Boelelaan 1111 Amsterdam1081 HV Netherlands School of Computing Engineering and the Built Environment Edinburgh Napier University 10 Colinton Road EdinburghEH10 5DT United Kingdom Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui University Jiulong Road 111 Anhui Hefei230601 China

The traveling salesperson problem (TSP) is one of the most iconic hard optimization tasks. With frequency fitness assignment (FFA), a new approach to optimization has recently been proposed: instead of directing the search towards better solutions, the optimization process prefers those with rarely encountered objective values. FFA can be plugged into existing algorithms and can improve their performance on some problems that are hard for them. However, problems like the TSP, where the number of possible objective values (here: tour lengths) is high, cause the performance of FFA-based algorithms to deteriorate. We choose 56 symmetric instances from TSPLib as a testbed for our experiments. We plug FFA both into the (1+1) EA and simulated annealing (SA) and obtain the FEA and FSA, respectively. We find that the resulting FEA substantially outperforms the (1+1) EA and that FSA can reach the global optima more often than SA within our computational budget. We propose two ways to hybridize FFA and traditional optimization in a round-robin scheme where objective function evaluations are alternatingly assigned to both components. The FFA-based parts of the algorithms sometimes send a solution over to the traditional-optimization-based ones. This can lead to further performance improvements. In particular, the hybrids combining simulated annealing with the FEA solve most problems and find close-to-optimal solutions on other instances. Even for several instances with many possible different objective values, which are therefore not suited to FFA, hybrids combining FFA and traditional objective-guided optimization algorithms can offer substantial performance improvements. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Simulated annealing

来源：评论

学校读者我要写书评

暂无评论

BroadCAM: Outcome-agnostic Class Activation Mapping for Small-scale Weakly Supervised Applications

arXiv

引用

arXiv 2023年

作者： Lin, Jiatai Han, Guoqiang Xu, Xuemiao Liang, Changhong Wong, Tien-Tsin Chen, C.L. Philip Liu, Zaiyi Han, Chu The School of Computer Science and Engineering South China University of Technology Guangdong Guangzhou510006 China Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangzhou510080 China Southern Medical University Guangzhou510080 China School of Computer Science and Engineering South China University of Technology State Key Laboratory of Subtropical Building Science Ministry of Education Key Laboratory of Big Data and Intelligent Robot Guangdong Provincial Key Lab of Computational Intelligence and Cyberspace Information China Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong

Class activation mapping (CAM), a visualization technique for interpreting deep learning models, is now commonly used for weakly supervised semantic segmentation (WSSS) and object localization (WSOL). It is the weighted aggregation of the feature maps by activating the high class-relevance ones. Current CAM methods achieve it relying on the training outcomes, such as predicted scores (forward information), gradients (backward information), etc. However, when with small-scale data, unstable training may lead to less effective model outcomes and generate unreliable weights, finally resulting in incorrect activation and noisy CAM seeds. In this paper, we propose an outcome-agnostic CAM approach, called BroadCAM, for small-scale weakly supervised applications. Since broad learning system (BLS) is independent to the model learning, BroadCAM can avoid the weights being affected by the unreliable model outcomes when with small-scale data. By evaluating BroadCAM on VOC2012 (natural images) and BCSS-WSSS (medical images) for WSSS and OpenImages30k for WSOL, BroadCAM demonstrates superior performance than existing CAM methods with small-scale data (less than 5%) in different CNN architectures. It also achieves SOTA performance with large-scale training data. Extensive qualitative comparisons are conducted to demonstrate how BroadCAM activates the high class-relevance feature maps and generates reliable CAMs when with small-scale training data. © 2023, CC BY.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Predicting Hand Injury Severity Using Bayesian Networks in Cloud-Based Emergency Medicine

Predicting Hand Injury Severity Using Bayesian Networks in C...

引用

International Conference on Communications and Signal Processing

作者： S Yuvaraj K. Thinakaran S Porandla Srinivas S. Sivakumar M.V Ishwarya P. Thirumaraiselvan Department of Electronics and Communication Engineering Faculty of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai Tamil Nadu India Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai Tamil Nadu India Department of Information Technology Malla Reddy Institute of Engineering and Technology Hyderabad Telangana India Department of Computer Science Thanthai Hans Roever College (Autonomous) Perambalur Tamil Nadu India Artificial Intelligence and Data science Department Agni College of Technology Thazhambur Tamil Nadu India Department of Electronics and Communication Engineering Adhiparasakthi Engineering College Melmaruvathur Tamil Nadu India

ISBN: (数字)9798350353068

ISBN: (纸本)9798350353075

Predicting the severity of hand injuries in emergency care is the focus of this study, which suggests using cloud-based Bayesian networks. It is crucial to have accurate severity prediction in order to treat emergency hand injuries promptly since they are prevalent. By using the scalability and flexibility of cloud computing, this method constructs a Bayesian severity assessment network with patient data, injury characteristics, and historical medical records. The model’s training data comes from a massive database of hand injury cases with different results. Emergency medical professionals may use real-time parameter modifications to strengthen and adapt the Bayesian network. Through the use of the cloud, healthcare providers may simply exchange data and reach consensus. It demonstrates that, compared to other well-established methods of evaluating hand injury severity, the model performs better in making predictions. The suggested method improves severity forecasts and advances cloud-based emergency care. The method for emergency hand injury care uses Bayesian networks in the cloud to improve scalability and accessibility. The findings show that the data-driven and efficient hand injury severity prediction method might transform emergency care by improving patient outcomes.

关键词： Cloud computing Adaptation models Scalability Training data Medical services Transforms Predictive models

来源：评论

学校读者我要写书评

暂无评论

Cross-Domain Animal Pose Estimation with Skeleton Anomaly-Aware Learning

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Han, Le Chen, Kaixuan Zhao, Lei Jiang, Yangbo Wang, Pengfei Zheng, Nenggan Zhejiang Hangzhou310007 China Zhejiang University College of Computer Science and Technology Zhejiang Hangzhou310007 China Zhejiang University State Key Laboratory of Blockchain and Data Security Zhejiang Hangzhou310007 China Institute of Blockchain and Data Security. China Zhejiang University School of Software Technology Ningbo China Zhejiang University State Key Lab of Brain-Machine Intelligence Hangzhou310007 China Zhejiang Provincial Government ZJU Collaborative Innovation Center for Artificial Intelligence by MOE Hangzhou310007 China Bengbu University School of Computer and Information Engineering Bengbu233030 China

Animal pose estimation is often constrained by the scarcity of annotations and the diversity of scenarios and species. The pseudo-label generation based unsupervised domain adaptation paradigm, which discriminates the predicted keypoints of unlabeled data based on the skeleton position consistency, has demonstrated effectiveness for such problems. However, existing methods generate pseudo-labels with massive false positives, because they cannot effectively distinguish sample pairs with the same errors. In this study, we propose a cross-domain animal pose estimation model from a novel perspective of skeleton anomaly learning. We construct a graph contrastive learning mechanism to acquire the skeleton anomaly-aware knowledge, which enables the generation of accurate pseudo-labels for target domain and imposes graph constraint on unlabeled data. And a skeleton anomaly-feedback based domain adaptation framework is designed to facilitate implicit alignment of object-specific features and joint training of cross-domain. Besides, we propose a novel rat pose dataset named UDARP-9.4K to address the gap of small-sized animal pose datasets encompassing diverse experimental scenarios. The related datasets are reviewed and evaluated in detail. Extensive experiments are conducted on UDARP-9.4K and two public datasets to demonstrate the superiority of the proposed model in cross-scenarios and cross-species animal pose estimation tasks. Further analysis reveals the effectiveness of the proposed model for skeleton structure feature learning. The UDARP-9.4K dataset is available here. © 1991-2012 IEEE.

关键词： Invertebrates

来源：评论

学校读者我要写书评

暂无评论

Development and Validation of a Prognostic Risk Score System for COVID-19 Inpatients:A Multi-Center Retrospective Study in China

引用

engineering 2022年第1期8卷 116-121页

作者： Ye Yuan Chuan Sun Xiuchuan Tang Cheng Cheng Laurent Mombaerts Maolin Wang Tao Hu Chenyu Sun Yuqi Guo Xiuting Li Hui Xu Tongxin Ren Yang Xiao Yaru Xiao Hongling Zhu Honghan Wu Kezhi Li Chuming Chen Yingxia Liu Zhichao Liang Zhiguo Cao Hai-Tao Zhang Ioannis Ch.Paschaldis Quanying Liu Jorge Goncalves Qiang Zhong Li Yan School of Artificial Intelligence and Automation Huazhong University of Science and TechnologyWuhan 430074China School of Mechanical Science and Engineering Huazhong University of Science and TechnologyWuhan 430074China Luxembourg Centre for Systems Biomedicine University of LuxembourgBelval L-4367Luxembourg Department of Plant Sciences University of CambridgeCambridge CB21TNUK Department of Emergency Tongji HospitalTongji Medical CollegeHuazhong University of Science and TechnologyWuhan 430030China AMITA Health Saint Joseph Hospital Chicago ChicagoIL 60657USA Department of Anesthesiology Tongji HospitalTongji Medical CollegeHuazhong University of Science and TechnologyWuhan 430030China Huazhong University of Science and Technology-Wuxi Research Institute Wuxi 214174China Division of Cardiology Department of Internal MedicineTongji HospitalTongji Medical CollegeHuazhong University of Science and TechnologyWuhan 430030China Institute of Health Informatics University College LondonLondon NW12DAUK Department of Biomedical Engineering Southern University of Science and TechnologyShenzhen 518055China Department of Infectious Diseases Shenzhen Key Laboratory of Pathogenic Microbiology and ImmunologyNational Clinical Research Center for Infectious DiseaseThe Third People’s Hospital of Shenzhen(Second Hospital Affiliated with the Southern University of Science and Technology)Shenzhen 518055China Department of Electrical and Computer Engineering&Division of Systems Engineering&Department of Biomedical Engineering Boston UniversityBostonMA 02215USA

Coronavirus disease 2019(COVID-19)has become a worldwide *** patients of COVID-19 suffer from a high mortality rate,motivating the development of convenient and practical methods that allow clinicians to promptly identify high-risk ***,we have developed a risk score using clinical data from 1479 inpatients admitted to Tongji Hospital,Wuhan,China(development cohort)and externally validated with data from two other centers:141 inpatients from Jinyintan Hospital,Wuhan,China(validation cohort 1)and 432 inpatients from The Third People’s Hospital of Shenzhen,Shenzhen,China(validation cohort 2).The risk score is based on three biomarkers that are readily available in routine blood samples and can easily be translated into a probability of *** risk score can predict the mortality of individual patients more than 12 d in advance with more than 90%accuracy across all ***,the Kaplan-Meier score shows that patients can be clearly differentiated upon admission as low,intermediate,or high risk,with an area under the curve(AUC)score of *** summary,a simple risk score has been validated to predict death in patients infected with severe acute respiratory syndrome coronavirus 2(SARS-CoV-2);it has also been validated in independent cohorts.

关键词： COVID-19 Risk score Mortality risk prediction

来源：评论

学校读者我要写书评

暂无评论

Offline Equilibrium Finding

arXiv

引用

arXiv 2022年

作者： Li, Shuxin Wang, Xinrun Zhang, Youzhi Černý, Jakub Li, Pengdeng Chan, Hau An, Bo School of Computer Science and Engineering Nanyang Technological University Singapore Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences China University of Nebraska-Lincoln United States

Offline reinforcement learning (offline RL) is an emerging field that has recently begun gaining attention across various application domains due to its ability to learn strategies from earlier collected datasets. Offline RL proved very successful, paving a path to solving previously intractable real-world problems, and we aim to generalize this paradigm to a multiplayer-game setting. To this end, we introduce a problem of offline equilibrium finding (OEF) and construct multiple types of datasets across a wide range of games using several established methods. To solve the OEF problem, we design a model-based framework that can directly apply any online equilibrium finding algorithm to the OEF setting while making minimal changes. The three most prominent contemporary online equilibrium finding algorithms are adapted to the context of OEF, creating three model-based variants: OEF-PSRO and OEF-CFR, which generalize the widely-used algorithms PSRO and Deep CFR to compute Nash equilibria (NEs), and OEF-JPSRO, which generalizes the JPSRO to calculate (Coarse) Correlated equilibria ((C)CEs). We also combine the behavior cloning policy with the model-based policy to further improve the performance and provide a theoretical guarantee of the solution quality. Extensive experimental results demonstrate the superiority of our approach over offline RL algorithms and the importance of using model-based methods for OEF problems. We hope our work will contribute to advancing research in large-scale equilibrium finding. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A data-Driven Near-Optimization Approach for Smart Parking Management Platforms 18

A Data-Driven Near-Optimization Approach for Smart Parking M...

引用

18th IEEE International Conference on Networking, Sensing and Control, ICNSC 2021

作者： Bai, Mingyan Zhong, Shenghua Yan, Pengyu Chen, Zhibin Zhang, Zhixian University of Electronic Science and Technology of China School of Management and Economics Chengdu China Chengdu Communication Investment Group Intelligent Parking Industrial Development Company Ltd Chengdu China NYU Shanghai Division of Engineering and Computer Science Center for Data Science and Artificial Intelligence Shanghai China China Telecom Research Institute Shanghai China

ISBN: (纸本)9781665440486

This paper addresses an allocation optimization problem of parking slots in a real-time parking reservation platform in which parking demands randomly show up. For each decision period of a finite horizon, the reservation platform allocates the available parking slots at hand for the random demands with different types specified by the parking duration to maximize the total revenue over the whole horizon. This paper presents a real-time reservation framework and formulates the problem as a stochastic programming model, considering the different type demands with unknown probability distributions. We propose a data-driven near-optimization approach entitled a two-sample average approximation (2-SAA) to determine the allocation scheme over the horizon. In the 2-SAA approach, the confidence interval of the total revenue is established by the out-of-sample resampling method. The results of the numerical experiment validate the performance of the proposed 2-SAA algorithm. © 2021 IEEE.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

DARWIN 1.5: Large Language Models as Materials science Adapted Learners

arXiv

引用

arXiv 2024年

作者： Xie, Tong Wan, Yuwei Liu, Yixuan Zeng, Yuchen Wang, Shaozhou Zhang, Wenjie Grazian, Clara Kit, Chunyu Ouyang, Wanli Zhou, Dongzhan Hoex, Bram School of Photovoltaic and Renewable Energy Engineering University of New South Wales KensingtonNSW Australia GreenDynamics SydneyNSW Australia Department of Linguistics and Translation City University of Hong Kong Hong Kong Department of Computer Science University of Wisconsin-Madison MadisonWI United States Shanghai Artificial Intelligence Laboratory Shanghai China School of Computer Science and Engineering University of New South Wales KensingtonNSW Australia School of Mathematics and Statistics University of Sydney CamperdownNSW Australia DARE ARC Training Centre in Data Analytics for Resources and Environments CamperdownNSW Australia

Materials discovery and design aim to find compositions and structures with desirable properties over highly complex and diverse physical spaces. Traditional solutions, such as high-throughput simulations or machine learning, often rely on complex descriptors, which hinder generalizability and transferability across different material systems. Moreover, These descriptors may inadequately represent macro-scale material properties, which are influenced by structural imperfections and compositional variations in real-world samples, thus limiting their practical applicability. To address these challenges, we propose DARWIN 1.5, the largest open-source large language model tailored for materials science. By leveraging natural language as input, DARWIN eliminates the need for task-specific descriptors and enables a flexible, unified approach to material property prediction and discovery. Our approach integrates 6M material domains papers and 21 experimental datasets from 49,256 materials across modalities while enabling cross-task knowledge transfer. The enhanced model achieves up to 59.1% improvement in prediction accuracy over the base LLaMA-7B architecture and outperforms SOTA machine learning approaches across 8 materials design tasks. These results establish LLMs as a promising foundation for developing versatile and scalable models in materials science. © 2024, CC BY.

关键词： Materials properties

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：