检索结果-南通市图书馆

Estimating differential quantities from point cloud based on a linear fitting of normal vectors

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science in China(Series F) 2009年第3期52卷 431-444页

作者： CHENG zhangLin zhang xiaopeng Sino-French Laboratory LLAMA National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of SciencesBeijing 100080 China

Estimation of differential geometric properties on a discrete surface is a fundamental work in computer graphics and computer vision. In this paper, we present an accurate and robust method for estimating differential quantities from unorganized point cloud. The principal curvatures and principal directions at each point are computed with the help of partial derivatives of the unit normal vector at that point, where the normal derivatives are estimated by fitting a linear function to each component of the normal vectors in a neighborhood. This method takes into account the normal information of all neighboring points and computes curvatures directly from the variation of unit normal vectors, which improves the accuracy and robustness of curvature estimation on irregular sampled noisy data. The main advantage of our approach is that the estimation of curvatures at a point does not rely on the accuracy of the normal vector at that point, and the normal vectors can be refined in the process of curvature estimation. Compared with the state of the art methods for estimating curvatures and Darboux frames on both synthetic and real point clouds, the approach is shown to be more accurate and robust for noisy and unorganized point cloud data.

关键词： differential geometric properties point cloud normal fitting Weingarten matrix

What makes for uniformity for non-contrastive self-supervised learning?

在线全文

学校读者我要写书评

暂无评论

Science China(Technological Sciences) 2022年第10期65卷 2399-2408页

作者： WANG YinQuan zhang xiaopeng TIAN Qi Lü JinHu Key Laboratory of Systems and Control Academy of Mathematics and Systems ScienceChinese Academy of SciencesBeijing 100190China School of Mathematical Sciences University of Chinese Academy of SciencesBeijing 100049China Huawei Inc. Shenzhen 518128China School of Automation Science and Electrical Engineering State Key Laboratory of Software Development EnvironmentBeihang UniversityBeijing 100191China

Recent advances in self-supervised learning(SSL) have made remarkable progress, especially for contrastive methods that target pulling two augmented views of one image together and pushing the views of all other images away. In this setting, negative pairs play a key role in avoiding collapsed representation. Recent studies, such as those on bootstrap your own latent(BYOL)and SimSiam, have surprisingly achieved a comparable performance even without contrasting negative samples. However, a basic theoretical issue for SSL arises: how can different SSL methods avoid collapsed representation, and is there a common design principle? In this study, we look deep into current non-contrastive SSL methods and analyze the key factors that avoid *** achieve this goal, we present a new indicator of uniformity metric and study the local dynamics of the indicator to diagnose collapses in different scenarios. Moreover, we present some principles for choosing a good predictor, such that we can explicitly control the optimization process. Our theoretical analysis result is validated on some widely used benchmarks spanning differentscale datasets. We also compare recent SSL methods and analyze their commonalities in avoiding collapses and some ideas for future algorithm designs.

关键词： contrastive learning self-supervised learning representation uniformity dynamics

同方期刊数据库

Easy modeling of realistic trees from freehand sketches

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2012年第6期6卷 756-768页

作者： Jia LIU Zhiguo JIANG Hongjun LI xiaopeng zhang Image Processing Center School of AstronauticsBeihang UniversityBeijing 100191China NLPR-LIAMA Institute of AutomationChinese Academy of SciencesBeijing 100190China College of Science Beijing Forestry UniversityBeijing 100083China

Creating realistic 3D tree models in a convenient way is a challenge in game design and movie making due to diversification and occlusion of tree structures. Current sketch-based and imagebased approaches for fast tree modeling have limitations in effect and speed, and they generally include complex parameter adjustment, which brings difficulties to novices. In this paper, we present a simple method for quickly generating various 3D tree models from freehand sketches without parameter adjustment. On two input images, the user draws strokes representing the main branches and crown silhouettes of a tree. The system automatically produces a 3D tree at high speed. First, two 2D skeletons are built from strokes, and a 3D tree structure resembling the input sketches is built by branch retrieval from the 2D skeletons. Small branches are generated within the sketched 2D crown silhouettes based on self-similarity and angle restriction. This system is demonstrated on a variety of examples. It maintains the main features of a tree： the main branch structure and crown shape, and can be used as a convenient tool for tree simulation and design.

关键词： virtual reality tree modeling sketch

Bi-material Topology Optimization Using Analysis Mesh-Independent Point-Wise Density Interpolation

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Acta Mechanica Solida Sinica 2019年第6期32卷 698-712页

作者： Xiaoyu Suo Zhan Kang xiaopeng zhang Yaguang Wang State Key Laboratory of Structural Analysis for Industrial Equipment Dalian University of TechnologyDalian 116024China The 41st Institute of the Sixth Academy of China Aerospace Science&Industry Corporation Hohhot 010010China

This paper extends the independent point-wise density interpolation to the bimaterial to pology optimization to improve the structural static or dynamic proper *** contras t to the conventional elemental density-based topology optimization approaches,this method employs an analysis-mesh-separated material density field discretization model to describe the topology evolution of bi-material structures within the design *** be specific,the density design variable points can be freely positioned,independently of the field points used for discretization of the displacement *** this means,a material interface description of relatively high quality can be achieved,even when unstructured finite element meshes and irregular-shaped elements are used in discretization of the analysis *** examples,regarding the minimum static compliance design and the maximum fundamental eigen-frequency design,are presented to demonstrate the validity and applicability of the proposed formulation and numerical *** is shown that this method is free of numerical difficulties such as checkerboard patterns and the“islanding”phenomenon.

关键词： Topology optimization Bi-material Independent point-wise density interpolation Topology description Material interface Dynamic topology optimization

A MATLAB code for the material-field series-expansion topology optimization method

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Mechanical Engineering 2021年第3期16卷 607-622页

作者： Pai LIU xiaopeng zhang Yangjun LUO Yi YAN State Key Laboratory of Structural Analysis for Industrial Equipment Dalian University of TechnologyDalian 116024China School of Aeronautics and Astronautics Dalian University of TechnologyDalian 116024China

This paper presents a MATLAB implementation of the material-field series-expansion(MFSE)topology optimization *** MFSE method uses a bounded material field with specified spatial correlation to represent the structural *** the series-expansion method for bounded fields,this material field is described with the characteristic base functions and the corresponding *** with the conventional density-based method,the MFSE method decouples the topological description and the finite element discretization,and greatly reduces the number of design variables after dimensionality *** features of this method include inherent control on structural topological complexity,crisp structural boundary description,mesh independence,and being free from the checkerboard *** the focus on the implementation of the MFSE method,the present MATLAB code uses the maximum stiffness optimization problems solved with a gradientbased optimizer as *** MATLAB code consists of three parts,namely,the main program and two subroutines(one for aggregating the optimization constraints and the other about the method of moving asymptotes optimizer).The implementation of the code and its extensions to topology optimization problems with multiple load cases and passive elements are discussed in *** code is intended for researchers who are interested in this method and want to get started with it *** can also be used as a basis for handling complex engineering optimization problems by combining the MFSE topology optimization method with non-gradient optimization algorithms without sensitivity information because only a few design variables are required to describe relatively complex structural topology and smooth structural boundaries using the MFSE method.

关键词： MATLAB implementation topology optimization material-field series-expansion method bounded material field dimensionality reduction

Statistical learning based facial animation

在线全文

学校读者我要写书评

暂无评论

Journal of Zhejiang University-Science C(Computers and Electronics) 2013年第7期14卷 542-550页

作者： Shibiao XU Guanghui MA Weiliang MENG xiaopeng zhang National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences

To synthesize real-time and realistic facial animation, we present an effective algorithm which combines image- and geometry-based methods for facial animation simulation. Considering the numerous motion units in the expression coding system, we present a novel simplified motion unit based on the basic facial expression, and construct the corresponding basic action for a head model. As image features are difficult to obtain using the performance driven method, we develop an automatic image feature recognition method based on statistical learning, and an expression image semi-automatic labeling method with rotation invariant face detection, which can improve the accuracy and efficiency of expression feature identification and training. After facial animation redirection, each basic action weight needs to be computed and mapped automatically. We apply the blend shape method to construct and train the corresponding expression database according to each basic action, and adopt the least squares method to compute the corresponding control parameters for facial animation. Moreover, there is a pre-integration of diffuse light distribution and specular light distribution based on the physical method, to improve the plausibility and efficiency of facial rendering. Our work provides a simplification of the facial motion unit, an optimization of the statistical training process and recognition process for facial animation, solves the expression parameters, and simulates the subsurface scattering effect in real time. Experimental results indicate that our method is effective and efficient, and suitable for computer animation and interactive applications.

关键词： Facial animation Motion unit Statistical learning Realistic rendering Pre-integration

Thermoviscoelastic Modeling Approach for Predicting the Recovery Behaviors of Thermally Activated Amorphous Shape Memory Polymers

在线全文

学校读者我要写书评

暂无评论

Transactions of Nanjing University of Aeronautics and Astronautics 2019年第5期36卷 798-807页

作者： GU Jianping FANG Changqing SUN Huiyu zhang xiaopeng Jiangsu Key Laboratory of Advanced Structural Materials and Application Technology School of Materials Science and EngineeringNanjing Institute of TechnologyNanjing 211167P.R.China Shanghai Space Propulsion Technology Research Institute Shanghai 201109P.R.China State Key Laboratory of Mechanics and Control of Mechanical Structures Nanjing University of Aeronautics and AstronauticsNanjing 210016P.R.China

A thermoviscoelastic modeling approach is developed to predict the recovery behaviors of the thermally activated amorphous shape memory polymers(SMPs)based on the generalized finite deformation viscoelasticity *** this paper,a series of moduli and relaxation times of the generalized Maxwell model is estimated from the stress relaxation master curve by using the nonlinear regression(NLREG)*** that the amorphous SMPs are approximately incompressible isotropic elastomers in the rubbery state,the hyperelastic response of the materials is well modeled with a hyperelastic model in Ogden *** addition,the Williams-Landel-Ferry(WLF)equation is used to describe the horizontal shift factor obtained with time-temperature superposition principle(TTSP).The finite element simulations show good agreement with the experimental thermomechanical ***,the possibility of developing a temperature-responsive intravascular stent with the SMP studied here is investigated in terms of its thermomechanical ***,it can be concluded that the model has good prediction capabilities for the recovery behaviors of amorphous SMPs.

关键词： shape memory polymers(SMPs) thermoviscoelastic modeling approach finite deformation recovery behavior

Application of 3D stereoscopic visualization technology in casting aspect

在线全文

学校读者我要写书评

暂无评论

China Foundry 2014年第4期11卷 308-313页

作者： Kang Jinwu zhang xiaopeng zhang Chi Liu Baicheng Key Laboratory for Advanced Materials Processing Technology Ministry of EducationSchool of Materials Science and EngineeringTsinghua University

3D stereoscopic visualization technology is coming into more and more common use in the field of entertainment,and this technology is also beginning to cut a striking figure in casting industry and scientific *** history,fundamental principle,and devices of 3D stereoscopic visualization technology are reviewed in this *** authors’research achievements on the 3D stereoscopic visualization technology in the modeling and simulation of the casting process are *** technology can be used for the observation of complex 3D solid models of castings and the simulated results of solidification processes such as temperature,fluid flow,displacement,stress strain and microstructure,as well as the predicted defects such as shrinkage/porosity,cracks,and *** can also be used for other areas relating to 3D models,such as assembling of dies,cores,*** cases are given to compare the illustration of simulated results by traditional images and red-blue 3D stereoscopic *** spatial shape is observed better by the new *** prospect of3D stereoscopic visualization in the casting aspect is discussed as *** need for aided-viewing devices is still the most prominent problem of 3D stereoscopic visualization ***,3D stereoscopic visualization represents the tendency of visualization technology in the future;and as the problem is solved in the years ahead,great breakthroughs will certainly be made for its application in casting design and modeling and simulation of the casting processes.

关键词： 3D stereoscopic visualization technology casting modeling and simulation observation and analysis

A comparative study of Mn/CeO_2,Mn/ZrO_2 and Mn/Ce-ZrO_2 for low temperature selective catalytic reduction of NO with NH_3 in the presence of SO_2 and H_2O

在线全文

学校读者我要写书评

暂无评论

Journal of Environmental Sciences 2013年第4期25卷 791-800页

作者： Boxiong Shen xiaopeng zhang Hongqing Ma Yan Yao Ting Liu College of Environmental Science and Engineering Nankai University

Ce-ZrO2 is a widely used three-way catalyst support. Because of the large surface area and excellent redox quality, Ce-ZrO2 may have potential application in selective catalytic reduction （SCR） systems. In the present work, Ce-ZrO2 was introduced into a low-temperature SCR system and CeO2 and ZrO2 supports were also introduced to make a contrastive study. Mn/CeO2, Mn/ZrO2 and Mn/Ce-ZrO2 were prepared by impregnating these supports with Mn（NO3）2 solution, and have been characterized by N2-BET, XRD, TPR, TPD, XPS, FT-IR and TG. The activity and resistance to SO2 and H2O of the catalysts were investigated. Mn/Ce-ZrO2 and Mn/CeO2 were proved to have better low-temperature activities than Mn/ZrO2, and yielded 98.6% and 96.8% NO conversion at 180℃, respectively. This is mainly because Mn/Ce-ZrO2 and Mn/CeO2 had higher dispersion of manganese oxides, better redox properties and more weakly adsorbed oxygen species than Mn/ZrO2. In addition, Mn/Ce-ZrO2 showed a good resistance to SO2 and H2O and presented 87.1% NO conversion, even under SO2 and H2O treatment for 6 hours, and the activity of Mn/Ce-ZrO2 was almost restored to its original level after cutting off the injection of SO2 and H2O. This was due to the weak water absorption and weak sulfation process on the surface of the catalyst.

关键词： low temperature selective catalytic reduction manganese Ce-ZrO2 SO2 poisoning