检索结果-内蒙古大学图书馆

46th Annual Conference of the German Association for Pattern Recognition, DAGM-GCPR 2024

作者： Glaszner, Lukas Zach, Martin Institute of Computer Graphics and Vision Graz University of Technology Inffeldgasse 16/II Graz8010 Austria

ISBN: (纸本)9783031851803

Diffusion models have been successfully applied to many inverse problems, including MRI and CT reconstruction. Researchers typically re-purpose models originally designed for unconditional sampling without modifications. Using two different posterior sampling algorithms, we show empirically that such large networks are not necessary. Our smallest model, effectively a ResNet, performs almost as good as an attention U-Net on in-distribution reconstruction, while being significantly more robust towards distribution shifts. Furthermore, we introduce models trained on natural images and demonstrate that they can be used in both MRI and CT reconstruction, out-performing model trained on medical images in out-of-distribution cases. As a result of our findings, we strongly caution against simply re-using very large networks and encourage researchers to adapt the model complexity to the respective task. Moreover, we argue that a key step towards a general diffusion-based prior is training on natural images. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

FX-MAD: Frequency-Domain Explainability and Explainability-Driven Unsupervised Detection of Face Morphing Attacks

FX-MAD: Frequency-Domain Explainability and Explainability-D...

引用

2025 IEEE/CVF Winter Conference on Applications of computer Vision Workshops, WACVW 2025

作者： Huber, Marco Neto, Pedro C. Sequeira, Ana F. Damer, Naser Fraunhofer Institute for Computer Graphics Research Igd Darmstadt Germany Technische Universität Darmstadt Department of Computer Science Germany Inesc Tec Porto Portugal University of Porto Faculry of Engineering Portugal

ISBN: (纸本)9798331536626

Face recognition (FR) systems are vulnerable to morphing attacks, which refer to face images created by morphing the facial features of two different identities into one face image to create an image that can match both identities, allowing serious security breaches. In this work, we apply a frequency-based explanation method from the area of explainable face recognition to shine a light on how FR models behave when processing a bona fide or attack pair from a frequency perspective. In extensive experiments, we used two different state-of-the-art FR models and six different morphing attacks to investigate possible differences in behavior. Our results show that FR models rely differently on different frequency bands when making decisions for bona fide pairs and morphing attacks. In the following step, we show that this behavioral difference can be used to detect morphing attacks in an unsupervised setup solely based on the observed frequency-importance differences in a generalizable manner. © 2025 IEEE.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

Recognition and evaluation of cutaneous condition through assorted artificial intelligence reliant algorithms

引用

International Journal of Information Technology (Singapore) 2025年 1-13页

作者： Mishra, Manmohan Yadav, Ajay Kumar Mazumdar, Bireshwar Dass Gupta, Prashant K. Panwar, Arvind Bharadwaj, Shivam Department of Computer Application United Institute of Management Prayagraj India School of Computer Science & Engineering Technology Bennett University Plot Nos 8-11 TechZone II Uttar Pradesh Greater Noida 201310 India Galgotias University Plot No. 2 Yamuna Expy opposite Buddha International Circuit Sector 17A Uttar Pradesh Prayagraj 203201 India

Our skin is the hefty organ that envelops and shields body. It prevents us from numerous fatal and non fatal diseases. It is observed that due to bacteria or other causes of infection, skin faces certain minor or life threatening diseases. The most prioritized step toward restoring health is early illness signs identification. Identifying Cutaneous Condition from clinical images is one of the foremost challenges in medical image investigation. In the presented study we will enlighten the various Artificial Intelligence techniques falling under the categories of supervised machine learning including Probabilistic classifier (Naïve Bayes), Statistical algorithm (Logistic Regression), Ensemble learning (Random Decision Trees), Data analysis technique (Convolutional Neural Network) and Kernel approach (Support Vector Machine) to identify and classify the cutaneous condition appropriately so that corrective measures of skin treatment can be endow with. The proposed approach entails collecting images as input, preprocessing, segmenting, feature extraction and lastly applying the classification algorithms to derive the Cutaneous Condition categories. Additional trials are conducted using the different approaches as indicated and it was discovered from the suggested tests that the Convolutional Neural Network strategy yields the best results overall. The proposed model is trained, tested, and evaluated using the International Skin Imaging Collaboration (ISIC) 2019 challenge dataset and Human Against Machine with 10,000 training images (HAM10000) for the detection of manifold Cutaneous Condition. © Bharati Vidyapeeth's institute of computer Applications and Management 2025.

关键词： Artificial intelligence Confusion matrix Convolutional neural networks Cutaneous condition Image processing Machine learning

来源：评论

学校读者我要写书评

暂无评论

AI-based character generation for disease stories: a case study using epidemiological data to highlight preventable risk factors

引用

i-com 2025年第1期24卷 259-281页

作者： Mittenentzwei, Sarah Garrison, Laura A. Budich, Beatrice Lawonn, Kai Dockhorn, Alexander Preim, Bernhard Meuschke, Monique Department of Simulation and Graphics Otto-von-Guericke University of Magdeburg Universitätsplatz 2 Magdeburg39106 Germany Department of Informatics University of Bergen Thormøhlens Gate 55 Bergen5020 Norway Faculty of Mathematics and Computer Science University of Jena Ernst-Abbe-Platz 2 Jena07743 Germany Institute of Information Processing Leibniz University Hannover Welfengarten 1 Hannover30167 Germany

Data-driven storytelling has grown significantly, becoming prevalent in various fields, including healthcare. In medical narratives, characters are crucial for engaging audiences, making complex medical information accessible, and potentially influencing positive behavioral and lifestyle changes. However, designing characters that are both educational and relatable to effectively engage audiences is challenging. We propose a GenAI-assisted pipeline for character design in data-driven medical stories, utilizing Stable Diffusion, a deep learning text-to-image model, to transform data into visual character representations. This approach reduces the time and artistic skills required to create characters that reflect the underlying data. As a proof-of-concept, we generated and evaluated two characters in a crowd-sourced case study, assessing their authenticity to the underlying data and consistency over time. In a qualitative evaluation with four experts with knowledge in design and health communication, the characters were discussed regarding their quality and refinement opportunities. The characters effectively conveyed various aspects of the data, such as emotions, age, and body weight. However, generating multiple consistent images of the same character proved to be a significant challenge. This underscores a key issue in using generative AI for character creation: the limited control designers have over the output. © 2025 the author(s), published by De Gruyter, Berlin/Boston.

关键词： data-storytelling GenAI medicine healthcare character generation

来源：评论

学校读者我要写书评

暂无评论

SelfMAD: Enhancing Generalization and Robustness in Morphing Attack Detection via Self-Supervised Learning

arXiv

引用

arXiv 2025年

作者： Ivanovska, Marija Todorov, Leon Damer, Naser Jain, Deepak Kumar Peer, Peter Štruc, Vitomir Faculty of Electrical Engineering University in Ljubljana Slovenia Faculty of Computer and Information Science University in Ljubljana Slovenia Dalian University of Technology China Fraunhofer Institute for Computer Graphics Research Germany

With the continuous advancement of generative models, face morphing attacks have become a significant challenge for existing face verification systems due to their potential use in identity fraud and other malicious activities. Contemporary Morphing Attack Detection (MAD) approaches frequently rely on supervised, discriminative models trained on examples of bona fide and morphed images. These models typically perform well with morphs generated with techniques seen during training, but often lead to sub-optimal performance when subjected to novel unseen morphing techniques. While unsupervised models have been shown to perform better in terms of generalizability, they typically result in higher error rates, as they struggle to effectively capture features of subtle artifacts. To address these shortcomings, we present SelfMAD, a novel self-supervised approach that simulates general morphing attack artifacts, allowing classifiers to learn generic and robust decision boundaries without overfitting to the specific artifacts induced by particular face morphing methods. Through extensive experiments on widely used datasets, we demonstrate that SelfMAD significantly outperforms current state-of-the-art MADs, reducing the detection error by more than 64% in terms of EER when compared to the strongest unsupervised competitor, and by more than 66%, when compared to the best performing discriminative MAD model, tested in cross-morph settings. The source code for SelfMAD is available at https://***/LeonTodorov/SelfMAD. © 2025, CC BY-NC-ND.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

FX-MAD: Frequency-Domain Explainability and Explainability-Driven Unsupervised Detection of Face Morphing Attacks

FX-MAD: Frequency-Domain Explainability and Explainability-D...

引用

IEEE Winter Applications and computer Vision Workshops (WACVW)

作者： Marco Huber Pedro C. Neto Ana F. Sequeira Naser Damer Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany Department of Computer Science Technische Universität Darmstadt Germany INESC TEC Porto Portugal Faculry of Engineering University of Porto Portugal

ISBN: (数字)9798331536626

ISBN: (纸本)9798331536633

关键词： computer vision Correlation Face recognition Soft sensors Frequency-domain analysis Conferences Benchmark testing Security Facial features

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Adaptive Filter for Head Tracking in Augmented Reality (AR)-based Flight Simulators

引用

IEEE Transactions on computers 2025年

作者： Nwobodo, Onyeka J. Kuaban, Godlove Suila Nkemeni, Valery Wereszczynski, Kamil Cyran, Krzysztof A. Silesian University of Technology Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Gliwice Poland Polish Academy of Sciences Institute of Theoretical and Applied Informatics Poland University of Buea Laboratory of Electrical Engineering and Computing Faculty of Engineering and Technology Buea Cameroon

This paper presents a novel approach for head tracking in augmented reality (AR) flight simulators using an adaptive fusion of Kalman and particle filters. This fusion dynamically balances the strengths of both algorithms, leveraging Kalman filters for computational efficiency and particle filters for handling non-linearities based on real-time factors such as sensor noise and head movement patterns. Our method demonstrates superior tracking precision and reduced latency, making it particularly effective for immersive pilot training in AR-based flight simulations. While focused on flight simulation, our approach holds high potential for broader applications in other AR and virtual reality (VR) environments where precise, real-time head tracking is crucial. These results provide actionable design guidance for developers optimizing tracking systems in environments that require fast response times and high accuracy. Future extensions of this research could explore the generalization of our approach across diverse AR applications, including human-computer interaction (HCI), medical simulations, and gaming. © 1968-2012 IEEE.

关键词： Flight simulators

来源：评论

学校读者我要写书评

暂无评论

To the Point: From Dynamic Heatmap Video to Gaze Points 25

To the Point: From Dynamic Heatmap Video to Gaze Points

引用

Proceedings of the 2025 Symposium on Eye Tracking Research and Applications

作者： Beryl Gnanaraj Swetha Manivasagam Jaya Sreevalsan-Nair Graphics-Visualization-Computing Lab International Institute of Information Technology Bangalore Bangalore Karnataka India Department of Computer Science and Engineering PSG College of Technology Coimbatore Tamil Nadu India

来源：评论

学校读者我要写书评

暂无评论

Bigger Isn’t Always Better: Towards a General Prior for Medical Image Reconstruction

arXiv

引用

arXiv 2025年

作者： Glaszner, Lukas Zach, Martin Institute of Computer Graphics and Vision Graz University of Technology Inffeldgasse 16/II Graz8010 Austria Biomedical Imaging Group École polytechnique fédérale de Lausanne Lausanne1015 Switzerland Center for Biomedical Imaging Lausanne1015 Switzerland

Diffusion model have been successfully applied to many inverse problems, including MRI and CT reconstruction. Researchers typically re-purpose models originally designed for unconditional sampling without modifications. Using two different posterior sampling algorithms, we show empirically that such large networks are not necessary. Our smallest model, effectively a ResNet, performs almost as good as an attention U-Net on in-distribution reconstruction, while being significantly more robust towards distribution shifts. Furthermore, we introduce models trained on natural images and demonstrate that they can be used in both MRI and CT reconstruction, out-performing model trained on medical images in out-of-distribution cases. As a result of our findings, we strongly caution against simply re-using very large networks and encourage researchers to adapt the model complexity to the respective task. Moreover, we argue that a key step towards a general diffusion-based prior is training on natural images. © 2025, CC BY-SA.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Novel method for robust bilateral filtering point cloud denoising

引用

Alexandria Engineering Journal 2025年 127卷 573-585页

作者： Yang, Huan Wang, Wei Wang, Yue Wang, Peng College of Computer Science and Technology Chongqing University of Posts and Telecommunications No. 2 Chongwen Road Chongqing Nan'an400065 China School of Advanced Manufacturing Engineering Chongqing University of Posts and Telecommunications No. 2 Chongwen Road Chongqing Nan'an400065 China Sevnce Robotics Co. Ltd Building 7 Phase II Internet Industrial Park Chongqing Liangjiang401123 China Big Data Institute BaoShan University No. 16 Yuanzheng Road Longyang District Yunnan Baoshan678000 China

Bilateral filtering for point cloud denoising relies on the normal of the points. However, normals are not inherent attributes of point clouds but are estimated using algorithms such as principal component analysis (PCA). Due to inaccuracies in the normals, bilateral filtering may lead to unexpected results, such as smoothing in sharp regions. This study addressed this issue by improving the accuracy of normals. Initially, PCA was employed to estimate the initial normals of the point cloud. The points were then categorized into edge and planar points by fitting a three-dimensional sphere. Subsequently, iterative weighted PCA was applied to refine the normals of the edge points. Finally, the improved normals were integrated into bilateral filtering to achieve point cloud denoising. Additionally, a robust estimator was used instead of a Gaussian function to compute the weights in the bilateral filtering. To assess and compare the performance of this algorithm with related ones, the mean square angular error (MSAE) and angular error distribution (AE) were employed to evaluate the accuracy of normal estimation, while the mean square error (MSE) and signal-to-noise ratio (SNR) were employed to assess the denoising performance. Results indicate that the proposed method yielded the most accurate normals compared to PCA, Jet, and VCM normal estimation algorithms. Moreover, when compared to algebraic point set surfaces (APSS), robust implicit moving least squares (RIMLS), anisotropic weighted locally optimal projection (AWLOP), bilateral filtering, and guided filtering point cloud denoising algorithms, the proposed method consistently achieved the smallest MSE and the highest SNR in most cases on the dataset used in this study. © 2025 The Authors

关键词： Principal component analysis

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：