基于语义网的数字化家具模型本体设计与检索

针对传统检索方式在三维模型爆发性增长背景下出现的种种缺陷,以家具模型为切入点,对基于语义网和本体技术的三维创意素材模型的本体构建与检索进行研究设计,实现三维数字模型智能化检索。将本体技术和语义检索与三维模型结合,对不同类型的家具模型进行数字化描述、特征提取及要素分类,通过OWL本体描述语言创建本体,并将本体模型存入数据库从而形成模型素材本体库,根据语义规则构造可被机器理解的检索方式,为实现大众参与下的创新创意设计和产品快速原型设计打下基础。     

一种三维模型检索直方图比较算法

直方图方法常用于计算机视觉领域,好的直方图比较算法能够有效地提高检索的性能.但直方图比较算法在三维模型检索中的应用则较少得到研究.应用方向梯度直方图提取三维模型的直方图特征,然后给出一种直方图比较算法用于三维模型检索.实验结果表明：该直方图比较算法对某些种类的模型的检索性能比已有的算法更优.     

Support Vector Machine active learning for 3D model retrieval

In this paper, we present a novel Support Vector Machine active learning algorithm for effective 3D model retrieval using the concept of relevance feedback. The proposed method learns from the most informative objects which are marked by the user, and then creates a boundary separating the relevant models from irrelevant ones. What it needs is only a small number of 3D models labelled by the user. It can grasp the user's semantic knowledge rapidly and accurately. Experimental results showed that the proposed algorithm significantly improves the retrieval effectiveness. Compared with four state-of-the-art query refinement schemes for 3D model retrieval, it provides superior retrieval performance after no more than two rounds of relevance     

Watermarking on 3D mesh based on spherical wavelet transform

INTRODUCTIONTherapidgrowthofdigitalmediaoverInternetprovideseverybodywiththefacilityofeasyaccess,copy,editanddistributionofdigitalcontentssuchaselectronicdocuments,images,soundsandvideos.Thereisurgentdemandfortechniquestoprotectthecopyrightoftheoriginaldigitaldataandtopreventunauthorizedduplicationortampering.Digitalwatermarkingordatahidingisonesolutionforthecopyrightprotectionofdigitaldata.Digitalwatermarkingisaprocessbywhichauser-specifiedsignal(watermark)ishiddenorembeddedintoanothersi…     

Watermarking on 3D mesh based on spherical wavelet transform

In this paper we propose a robust watermarking algorithm for 3D mesh. The algorithm is based on spherical wavelet transform. Our basic idea is to decompose the original mesh into a series of details at different scales by using spherical wavelet transform; the watermark is then embedded into the different levels of details. The embedding process includes: global sphere parameterization, spherical uniform sampling, spherical wavelet forward transform, embedding watermark, spherical wavelet inverse transform, and at last resampling the mesh watermarked to recover the topological connectivity of the original model. Experiments showed that our algorithm can improve the capacity of the watermark and the robustness of watermarking against attacks.     

基于本征模型的三维模型特征提取与检索技术研究

对如何提高基于内容三维模型检索中的特征提取方法的效果进行了研究,提出了在特征提取过程中使用分类信息获得内容特征的方法．根据一类模型在内容上存在“大致类似,略有差异”的特点,提出采用有弹性的“本征模型”作为一类模型的代表．探索了本征模型特征的提取,较优本征模型的筛选,给出较为高效的基于本征模型的三维模型检索过程．采用多种评价指标对标准测试库Princeton Shape Benchmark的实验表明,该方法表现优异．     

基于网格和密度的数据流聚类算法

针对数据流的特点,提出了一种新的网格密度结合的GCTS算法.它采用了双层架构,在线层实现了网格密度参数的自设定,离线层以网格单元的重心为中心点,建立一个最大的子网格,使候选网格中的局部密集区域转化成了密集网格.最后使用最小生成树的算法生成进行聚类结果.提高了聚类效果.     

论数字校园环境下高校教学资源的整合运行策略

文章在调查分析20所"985工程"高校教学资源建设与整合情况的基础上,针对其中普遍存在问题,提出了教学资源整合应当以建设一个教学资源管理大平台为载体,综合运用中心整合、集成检索、资源导航等模式,对教学资源实施从数据、信息、知识以及服务等不同层次与角度的全方位、深层次整合,并与数字校园系统集成,为用户提供单点登录以及教学资源的个性化定制与推送服务的整合思路。文章还提出高校进行教学资源建设与整合工作的同时,需要制定相应的运行策略与管理机制,以保障教学资源建设与整合工作的正常运转与顺利进行。     

面向制造的板级电子电路产品多信息模型研究

Most electronics products use PCB to carry electronic circuits. This paper classifies information contained in PCB-based electronic circuits into several models: geometry model, physics model, performance model and function model. Based on this classification, a multi-information model of product is established. A composite model of product is also created based on object-orientation and characteristics of the product. The model includes a 3D geometry model, a physics model with integrated information that can be divided into microscopic and macroscopic information, a generalized performance model and a function model that are from top to bottom. Finally, a multi-unit analysis is briefly discussed.     

A novel dependency language model for information retrieval

This paper explores the application of term dependency in information retrieval (IR) and proposes a novel dependency retrieval model. This retrieval model suggests an extension to the existing language modeling (LM) approach to IR by introducing dependency models for both query and document. Relevance between document and query is then evaluated by reference to the Kullback-Leibler divergence between their dependency models. This paper introduces a novel hybrid dependency structure, which allows integration of various forms of dependency within a single framework. A pseudo relevance feedback based method is also introduced for constructing query dependency model. The basic idea is to use query-relevant top-ranking sentences extracted from the top documents at retrieval time as the augmented representation of query, from which the relationships between query terms are identified. A Markov Random Field (MRF) based approach is presented to ensure the relevance of the extracted sentences, which utilizes the association features between query terms within a sentence to evaluate the relevance of each sentence. This dependency retrieval model was compared with other traditional retrieval models. Experiments indicated that it produces significant improvements in retrieval effectiveness.     

A spherical parameterization approach based on symmetry analysis of triangular meshes

We present an efficient spherical parameterization approach aimed at simultaneously reducing area and angle distortions. We generate the final spherical mapping by independently establishing two hemisphere parameterizations. The essence of the approach is to reduce spherical parameterization to a planar problem using symmetry analysis of 3D meshes. Experiments and comparisons were undertaken with various non-trivial 3D models, which revealed that our approach is efficient and robust. In particular, our method produces almost isometric parameterizations for the objects close to the sphere.     

基于网络三维模型库的虚拟实验室研究

随着网络技术的发展和教学资源的扩大,改进和优化虚拟实验室体系结构,建立三维模型库,可以提高模型利用率并加快模型检索速度,实现虚拟教学资源的共享性和扩展性。初步探讨了基于网络模型库的虚拟实验室的体系结构,分析了三维模型库设计过程中需要考虑的几个问题,最后实现了一个简单的三维模型库原型系统,并应用在虚拟考察实验室中。     

土木工程课程虚拟仿真教学平台设计与应用

北京工业大学土木工程国家级虚拟仿真实验教学中心利用现代虚拟仿真最新技术,采取多种模式建立了虚拟仿真实验平台。以土木工程中基础实验课程教学方法为研究对象,提出基于虚拟仿真技术的土木工程实验实训课程的教学方案。依托真实系统和仿真实训中心,构建了土木工程应用领域广泛使用的钢筋混凝土简支梁实验、短柱实验、振动台实验等九个实训项目,同时构建了工程结构实验中心三维仿真系统,功能包括实验中心室内、外场景的整体3D展示以及大型实验设备和特色实验系统的3D模型、功能展示和试验仿真。     

动词体的球态语义学

本文介绍由塔斯基的立体几何导出的球态语义学,并将其应用于自然语言中的动词体现象.球态语义学特别适合应用于英语的进行体.这种方法有以下优点:(i)它扩展了区间式语义,并同时避免了其缺陷,(ii)它解决了未完成体难题,(iii)它的解决方法无需诉诸最终结果策略.逻辑方法一般被认为难于处理自然语言的动词体问题.基丁点的时间结构以及建立在该结构之上的经典普莱尔时态逻辑([18])太弱了.而基于区间的时态语义则缺乏足够的表达力,并且难以解释进行体([4,8]).本文给出一种新的基于球上整体-部分关系概念的模型和时态语义.这种球态语义学建基丁塔斯基1927年引入的立体儿何之上.与基丁点和基丁区间的语义不同,在球态语义学中很多动词体区分都能由统一的逻辑方法刻画.在一个由封闭球构成的论域中,可达关系由相切性概念给出.相应地,我们可定义外切、内切、外径、内径以及同心等基本概念.与区间式语义不同,球是论域的初始概念,球态语义学不是在时间段而是在球中对事件赋值.冈此,仅将时间区问作为初始概念而不承认其端点初始性的问题不复存在.英语中的进行体由球上的连续行动米刻画.行动是非终止的,只要球没有由外切相离.相应地,外切相离刻画动作完成.我们区分在均匀球和非均匀球中发生事什的整体-部分关系.非持续动作视为直径为零的同心球.球态语义学根据动作或执行完成的时刻米定义时间概念,其中不需要时间端点的概念.在保持与基于区间的时间模型类似的基础上,球态语义学暗示了一种关于可能世界的定性概念,并且它有利于解决时间的循环概念问题.     

A head in virtual reality: Development of a dynamic head and neck model

Advances in computer and interface technologies have made it possible to create three‐dimensional (3D) computerized models of anatomical structures for visualization, manipulation, and interaction in a virtual 3D environment. In the past few decades, a multitude of digital models have been developed to facilitate complex spatial learning of the human body. However, there is limited empirical evidence to guide the development and integration of effective computer models for teaching and learning. The purpose of this article is to describe the development of a dynamic head and neck model with flexible displays (2D, 3D, and stereoscopic 3D) and interactive control features that can be later used to design and test the efficacy of computer models as a means of improving student learning. The model was created using computer tomography scans of a human cadaver. Anatomical structures captured on the scans were segmented into discreet areas, and then reconstructed in three‐dimensions using specialized software. The final model consists of 70 distinct anatomical structures that can be displayed in 2D, 3D, or stereoscopic 3D. In 3D mode, a mouse can be used to actively and continuously interact with the model by manipulating viewer orientation, altering surface transparency, superimposing 2D scans with 3D reconstructions, removing or adding structures sequentially, and customizing animated scenes to show complex anatomical pathways or relationships. Anat Sci Educ 2: 294–301, 2009. © 2009 American Association of Anatomists.     

Load-carrying capacity and practical calculation method for hollow cylinder joints connected with H-shaped beams

A type of hollow cylinder joints connected with H-shaped beams is proposed for spatial structures. Based on von Mises yield criterion and perfect elasto-plasticity model, a series of finite element models of the joints is established, in which the effect of geometric nonlinearity is taken into account. Then mechanical behavior and load-carrying capacity of the joints were investigated, which were subjected to axial load, in- and out-plane bending moments, and their combinations. The results show that the ultimate loads of the joints are determined by the maximum displacement. Furthermore, the case of one joint connected with multiple beams was discussed. Experiments on a set of typical full-scale joints were conducted to understand the structural behavior and the failure mechanism of joint, and also to validate the finite element models. Finally, the practical calculation method was established through finite elements analysis (FEA) results and numerical fitting. The results show that the joints are more ductile and materially economical than welded hollow spherical joints, and the practical calculation method can provide a reference for direct design and the revision of relevant design codes.     

Optimized Method for Real-time Texture Reconstruction with RGB-D Camera

With the appearance of RGB-D camera, the field of three-dimensional(3D) reconstruction receives more and more attention. In this paper, we present an optimization approach to produce high-quality textured 3D models based on the real-time 3D reconstruction system.The resulting models of real-time texture reconstruction often suffer from blurring, ghosting, and other artifacts.Our approach addresses this texture quality problem using blur detection and an optimized weight function. Experimental results demonstrate that our approach can improve the quality of textured 3D models by reducing the blur and ghosts on the model surface.     

Ultimate Bearing Capacity of Hollow Spherical Joints Welded with Circular Pipes under Eccentric Loads

The National Swimming Center built for the 2008Beijing Olympic Games applies the polyhedron spatialframe structure, which is based on the foam physicstheory[1]. Its simple geometry is a square box of 170m×170 m×29 m. It can hold 17 000 people, theroof o…     

Maximizing modern distribution of complex anatomical spatial information: 3D reconstruction and rapid prototype production of anatomical corrosion casts of human specimens

Anatomical corrosion casts of human specimens are useful teaching aids. However, their use is limited due to ethical dilemmas associated with their production, their lack of perfect reproducibility, and their consumption of original specimens in the process of casting. In this study, new approaches with modern distribution of complex anatomical spatial information were explored to overcome these limitations through the digitalization of anatomical casts of human specimens through three‐dimensional (3D) reconstruction, rapid prototype production, and Web‐based 3D atlas construction. The corrosion cast of a lung, along with its associated arteries, veins, trachea, and bronchial tree was CT‐scanned, and the data was then processed by Mimics software. Data from the lung casts were then reconstructed into 3D models using a hybrid method, utilizing both “image threshold” and “region growing.” The fine structures of the bronchial tree, arterial, and venous network of the lung were clearly displayed and demonstrated their distinct relationships. The multiple divisions of bronchi and bronchopulmonary segments were identified. The 3D models were then uploaded into a rapid prototype 3D printer to physically duplicate the cast. The physically duplicated model of the lung was rescanned by CT and reconstructed to detect its production accuracy. Gross observation and accuracy detection were used to evaluate the duplication and few differences were found. Finally, Virtual Reality Modeling Language (VRML) was used to edit the 3D casting models to construct a Web‐based 3D atlas accessible through Internet Explorer with 3D display and annotation functions. Anat Sci Educ. © 2012 American Association of Anatomists.     

基于 DenseNet 的人脸脸型分类研究

脸型在三维人脸重建、人脸识别与检索等领域有着重要应用。针对脸型分类问题提出一种基于傅里叶描述符、三角形半径描述符和几何特征描述符的 DenseNet 网络脸型分类方法。通过主动形状模型方法定位得到人脸边缘轮廓点,分别使用傅里叶描述符、三角形半径描述符和几何特征描述符将轮廓点数值化为一维向量,并将其输入 DenseNet 网络进行训练,实现脸型分类。通过实验对比分析,该方法能够充分挖掘人脸形状信息,有效提高脸型分类准确率,同时避免旋转、尺度等影响。