检索结果-内蒙古大学图书馆

Global fusion of generalized camera model for efficient large-scale structure from motion

Science China(Information Sciences) 2017年第3期60卷 218-220页

作者： Hainan CUI Shuhan SHEN Zhanyi HU National Laboratory of Pattern Recognition Institute of AutomationChinese Academy of Sciences University of Chinese Academy of Sciences

Recently,interest has grown in building large-scale3D city models[1]from images captured by multicamera systems,such as cameras mounted on a car,e.g.,Google Street View,or on an unmanned aerial vehicle,e.g.,oblique ai... 详细信息

关键词： view

来源：评论

学校读者我要写书评

暂无评论

引用

作者： Duin, Robert P.W. Pçkalska, Elbieta Pattern Recognition Laboratory Delft University of Technology Delft Netherlands School of Computer Science University of Manchester United Kingdom

Human experts constitute pattern classes of natural objects based on their observed appearance. Automatic systems for pattern recognition may be designed on a structural description derived from sensor observations. Alternatively, training sets of examples can be used in statistical learning procedures. They are most powerful for vectorial object representations. Unfortunately, structural descriptions do not match well with vectorial representations. Consequently it is difficult to combine the structural and statistical approaches to pattern recognition. Structural descriptions may be used to compare objects. This leads to a set of pairwise dissimilarities from which vectors can be derived for the purpose of statistical learning. The resulting dissimilarity representation bridges thereby the structural and statistical approaches. The dissimilarity space is one of the possible spaces resulting from this representation. It is very general and easy to implement. This paper gives a historical review and discusses the properties of the dissimilarity space approaches illustrated by a set of examples on real world datasets. © 2011 Elsevier B.V. All rights reserved.

关键词： Vector spaces

来源：评论

学校读者我要写书评

暂无评论

Automatic Change Detection from SAR Images Based on Fuzzy Entropy Principle

引用

电子学报(英文版) 2007年第1期16卷 76-81页

作者： PAN Chunhong PRINET Veronique YANG Qing MA Songde National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing 100080 China

In this paper, we propose a general framework for automatic change detection of flooded areas in bi-temporal Synthetic aperture radar (SAR) imagery. Based on the fuzzy entropy principle, a single thresholding method is proposed. The difference image which is generated by subtracting, one SAR image from the other one, pixel by pixel, can be divided into two classes: changed and unchanged. Due to noise, this single threshold decision procedure applied in the difference image is prone to two kinds of errors: false positives and false negatives. To cope with this problem, the threshold method is performed again on the two separate classes. As a result, the difference image can be divided into changed, unchanged, and unlabelled classes. The unlabelled class contains most of the ambiguous pixels that are responsible for false positive and false negative errors. In order to identify the true category of the pixels in this class, we adopt a voting procedure to compute the saliency of each unlabelled candidate pixel by analyzing the region configuration in the local neighborhood. Finally, according to the radiometric characteristics of water bodies in SAR images, some verification measures are carried out to identify water changes.

关键词： SAR 雷达图象自动变化检测模糊熵原理

来源：评论

学校读者我要写书评

暂无评论

Topological magnetic corner states in two-dimensional ferrimagnetic metal-organic frameworks

引用

Physical Review B 2025年第7期111卷 075433-075433页

作者： Qing-Bo Liu Xiangyang Li Ziyang Yu Lun Xiong Hua-Hua Fu Hubei Key Laboratory of Optical Information and Pattern Recognition School of Optical Information and Energy Engineering School of Mathematics and Physics Wuhan Institute of Technology Wuhan 430073 China Key Laboratory of Materials Physics Institute of Solid State Physics Hefei Institutes of Physical Science (HFIPS) Chinese Academy of Sciences Hefei 230031 China School of Physics and Wuhan National High Magnetic Field Center Huazhong University of Science and Technology Wuhan 430074 China Institute for Quantum Science and Engineering Huazhong University of Science and Technology Wuhan Hubei 430074 China

Topological corner states have been extensively studied in two-dimensional (2D) topological insulators (TIs), yet they remain rarely reported in 2D magnetic metal-organic frameworks (MOFs). In this work, we uncover the nontrivial corner states protected by C2z and C4z symmetry in three 2D ferrimagnetic MOFs. In the ferrimagnetic Cr(ODZ)2 (DZ = 1.2.5-diazole) with a magnetic space group (MSG) 14.2.87, the occupied electronic bands contain the essential band representations (BRs) of Au@2b and Ag@2c with unconventional features, facilitating the formation of quantized fractional charge (e2) in both spin-up and spin-down channels under C2z symmetry protection. Even in the presence of spin-orbit coupling (SOC), these nontrivial corner states are well preserved. Additionally, the C4z symmetry protected corner states are also identified in two other types of ferrimagnetic MOFs, specifically, Cr(SDZ)2 and Cr(SeDZ)2 monolayers in MSG 100.5.840 in the presence of SOC, contributed by the following basic BRs: Au@2b+Ag@2d. In the former, the magnetic corner states only exist in the spin-up channel, whereas in the latter, they are present in both spin-up and spin-down channels. Our research findings offer an ideal material platform for exploring the application of magnetic corner states in a special topological device.

关键词： Ferrimagnetism

来源：评论

学校读者我要写书评

暂无评论

GaitC3I: Robust Cross-Covariate Gait recognition via Causal Intervention

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Wang, Jilong Hou, Saihui Guo, Xianda Huang, Yan Huang, Yongzhen Zhang, Tianzhu Wang, Liang University of Science and Technology of China School of Information Science and Technology Department of Automation Hefei230026 China Beijing Normal University School of Artificial Intelligence Beijing100875 China Watrix Technology Limited Co. Ltd Beijing100088 China Wuhan University School of Computer Science China Institute of Automation Chinese Academy of Sciences New Laboratory of Pattern Recognition Beijing100190 China

Cross-covariate gait recognition aims to analyze a pedestrian's gait to extract an identity representation that is invariant across varying covariates. However, prevailing methods that have achieved good results on controlled in-the-lab datasets often perform poorly on realistic datasets. In this work, we find a significant cause is that the widely used pairwise metric learning paradigm cannot correctly handle the relationship between samples from different covariate conditions. Even worse, it may yield harmful signals that inadvertently mislead models to focus on covariate-related features, particularly when covariate distributions vary across subjects. To address this issue, we propose a Cross-Covariate Causal Intervention (GaitC3I) framework, a unified causality-inspired approach aimed at enhancing the robustness of gait recognition across diverse conditions. Specifically, our method consists of two parts: (i) an effective causal intervention metric learning paradigm based on backdoor adjustment, which strategically mitigates spurious correlations induced by covariates, thus ensuring a more invariant gait representation;and (ii) an annotation-free selection strategy that progressively matches each positive sample with negative samples from similar covariate conditions at various granularities. We demonstrate the effectiveness of our GaitC3I through extensive evaluation on six popular gait datasets-Gait3D, GREW, OUMVLP, CASIA-B, CCPG, and CCGR-achieving substantial improvements. Our method not only outperforms existing state-of-the-art models but also provides a systematic solution to remove the spurious correlations in gait recognition. © 1991-2012 IEEE.

关键词： backdoor adjustment biometrics causal inference debiasing gait recognition Identification of persons

来源：评论

学校读者我要写书评

暂无评论

Searching for topological carbon allotropes and the possible nontrivial quasi-particle states in them

引用

Materials Today Physics 2025年 53卷

作者： Liu, Qing-Bo Xiong, Lun Yu, Ziyang Fu, Hua-Hua Hubei Key Laboratory of Optical Information and Pattern Recognition School of Optical Information and Energy Engineering School of Mathematics and Physics Wuhan Institute of Technology Wuhan430073 China School of Physics and Wuhan National High Magnetic Field Center Huazhong University of Science and Technology Wuhan430074 China Institute for Quantum Science and Engineering Huazhong University of Science and Technology Wuhan Hubei430074 China

Carbon, as one of the most abundant elements on earth, possesses numerous allotropes that display a diverse range of physical properties. In this study, we utilized ab initio calculations and symmetry analyses to investigate 703 carbon allotropes, leading to the discovery of 315 (32) topological phononic (electronic) materials with ideal nontrivial characteristics. The topological phonons encompass single, charge-two, three, and four Weyl phonons, as well as Dirac (Weyl) node-lines phonons. Topological electronic states include topological insulators, Dirac points (Type-II), triple nodal points, and more. To verify this significant discovery, we adopt several real carbon allotropes with uni (pbg) structure within space group (SG) No 178 (230) to showcase their topological characteristics. The uni structure exhibits a combination of single-pair Weyl phonons and one-nodal surface phonons, resulting in a single surface arc in the (100) surface Brillouin zone (BZ) and isolated double-helix surface states in the (110) surface BZ. In the topological semimetal pbg, perfect triple degenerate nodal points near the Fermi level are found, resulting in distinct surface states in the (001) and (110) surfaces BZ. This research not only significantly broadens our understanding of topological quasi-particle states in carbon allotropes, but also offer a valuable material platform for further exploration of topological electrons and phonons in light element materials. © 2025 Elsevier Ltd

关键词： Carbon

来源：评论

学校读者我要写书评

暂无评论

Survey on AI-Generated Media Detection: From Non-MLLM to MLLM

arXiv

引用

arXiv 2025年

作者： Zou, Yueying Li, Peipei Li, Zekun Huang, Huaibo Cui, Xing Liu, Xuannan Zhang, Chenghanyu He, Ran School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China School of Science Beijing University of Posts and Telecommunications Beijing100876 China School of Computer Science University of California Santa Barbara United States State Key Laboratory of Multimodal Artificial Intelligence Systems CASIA New Laboratory of Pattern Recognition CASIA School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100190 China

The proliferation of AI-generated media poses significant challenges to information authenticity and social trust, making reliable detection methods highly demanded. Methods for detecting AI-generated media have evolved rapidly, paralleling the advancement of Multimodal Large Language Models (MLLMs). Current detection approaches can be categorized into two main groups: Non-MLLM-based and MLLM-based methods. The former employs high-precision, domain-specific detectors powered by deep learning techniques, while the latter utilizes general-purpose detectors based on MLLMs that integrate authenticity verification, explainability, and localization capabilities. Despite significant progress in this field, there remains a gap in literature regarding a comprehensive survey that examines the transition from domain-specific to general-purpose detection methods. This paper addresses this gap by providing a systematic review of both approaches, analyzing them from single-modal and multimodal perspectives. We present a detailed comparative analysis of these categories, examining their methodological similarities and differences. Through this analysis, we explore potential hybrid approaches and identify key challenges in forgery detection, providing direction for future research. Additionally, as MLLMs become increasingly prevalent in detection tasks, ethical and security considerations have emerged as critical global concerns. We examine the regulatory landscape surrounding Generative AI (GenAI) across various jurisdictions, offering valuable insights for researchers and practitioners in this field. © 2025, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Determining pose of 3D objects with curved surfaces

引用

IEEE TRANSACTIONS ON pattern ANALYSIS AND MACHINE INTELLIGENCE 1996年第1期18卷 52-57页

作者： Chen, JL Stockman, GC Pattern Recognition and Image Processing Laboratory Michigan State University East Lansing MI USA

A method is presented for computing the pose of rigid 3D objects with arbitrary curved surfaces. Given an input image and a candidate object model and aspect, the method will verify whether or not the object is present acid if so, report pose parameters. The curvature method of Basri and Ullman is used to model points on the object rim, while stereo matching is used for internal edge points. The model allows an object edgemap to be predicted from pose parameters. Pose is computed via an iterative search for the best pose parameters. Heuristics are used so that matching can succeed in the presence of occlusion and artifact and without resorting to use of corresponding salient feature points. Bench tests and simulations show that the method almost always converges to ground truth pose parameters for a variety of objects and for a broad set of starting parameters in the same aspect.

关键词： pose determination 3D objects object tracking object modeling image matching recognition by alignment

来源：评论

学校读者我要写书评

暂无评论

Improving depth resolution of diffuse optical tomography with an exponential adjustment method based on maximum singular value of layered sensitivity

引用

Chinese Optics Letters 2008年第12期6卷 886-888页

作者：牛海晶郭平宋晓东蒋田仔 Image Processing and Pattern Recognition Laboratory Beijing Normal University School of Computer Science and Technology Beijing Institute of Technology School of Aerospace Science and Engineering Beijing Institute of Technology National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences

The sensitivity of diffuse optical tomography （DOT） imaging exponentially decreases with the increase of photon penetration depth, which leads to a poor depth resolution for DOT. In this letter, an exponential adjustment method （EAM） based on maximum singular value of layered sensitivity is proposed. Optimal depth resolution can be achieved by compensating the reduced sensitivity in the deep medium. Simulations are performed using a semi-infinite model and the simulation results show that the EAM method can substantially improve the depth resolution of deeply embedded objects in the medium. Consequently, the image quality and the reconstruction accuracy for these objects have been largely improved.

关键词： Diagnostic radiography Electroabsorption modulators Image quality Light absorption Medical imaging Optical tomography

来源：评论

学校读者我要写书评

暂无评论

One-vs-all training of prototype classifier for pattern classification and retrieval

One-vs-all training of prototype classifier for pattern clas...

引用

2010 20th International Conference on pattern recognition, ICPR 2010

作者： Liu, Cheng-Lin National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences 95 Zhongguancun East Road Beijing 100190 China

ISBN: (纸本)9780769541099

Prototype classifiers trained with multi-class classification objective are inferior in pattern retrieval and outlier rejection. To improve the binary classification (detection, verification, retrieval, outlier rejection) performance of prototype classifiers, we propose a one-vs-all training method, which enriches each prototype as a binary discriminant function with a local threshold, and optimizes both the prototype vectors and the thresholds on training data using a binary classification objective, the cross-entropy (CE). Experimental results on two OCR datasets show that prototype classifiers trained by the one-vs-allmethod is superior in both multi-class classification and binary classification. © 2010 IEEE.

关键词： Classification (of information)

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：