基于草图局部几何不变矩的图像检索方法

利用草图进行图像检索的难点在于对不同尺度、位置、旋转及形变图像的有效检索。为了更准确地识别并检索不同尺度、位置和旋转的图像,提出一种基于草图局部几何不变矩的图像检索方法（SBIRULGMI）。首先,利用图像的几何特征分别确定各图像的坐标系;然后,在生成的坐标系中对图像进行平均分块并计算各块的几何不变矩作为特征向量;接着,用改进的欧氏距离计算目标图像与数据库图像的相似度;最后,采用蚁群（ACO）算法对按照相似度排序后的检索结果进行优化。所提方法在MPEG-7 shape1 part B图像数据库的检索识别准确率比形状上下文（SC）、边缘分布直方图（EOH）、局部线性高波特征（GALIF）及MindFinder方法平均提高了17个百分点。实验结果表明该方法对不同平移、缩放和翻转的图像有较好的识别效果,对图像一定程度的旋转和形变具有更好的鲁棒性。     

Sketch-based image matching Using Angular partitioning

This work presents a novel method for image similarity measure, where a hand-drawn rough black and white sketch is compared with an existing data base of full color images (art works and photographs). The proposed system creates ambient intelligence in terms of the evaluation of nonprecise, easy to input sketched information. The system can then provide the user with options of either retrieving similar images in the database or ranking the quality of the sketch against a given standard, i.e., the original image model. Alternatively, the inherent pattern-matching capability of the system can be utilized to allow detection of distortion in any given real time-image sequences in vision-driven ambient intelligence applications. The proposed method can cope with images containing several complex objects in an inhomogeneous background. Two abstract images are obtained using strong edges of the model image and the morphologically thinned outline of the sketched image. The angular-spatial distribution of pixels in the abstract images is then employed to extract new compact and effective features using the Fourier transform. The extracted features are rotation and scale invariant and robust against translation. Experimental results from seven different approaches confirm the efficacy of the proposed method in both the retrieval performance and the time required for feature extraction and search.     

草绘3维人体建模的模板形变方法

目的 采用草绘交互方式直接构造3维人体模型是当前人体建模研究的重要课题之一.提出一种草绘3维人体建模的模板形变方法.方法 针对输入的草图,首先,采用关节点定位方法获取草图中的人体关节点,根据人体结构学约束识别人体骨架结构,通过解析人体轮廓草图获取人体草图特征;其次,通过骨架模板和外观轮廓模板形变,将草图特征映射到3维人体模型,实现3维人体建模.结果 草图解析方法能有效提取草图特征,通过模板形变方法生成3维人体模型,并在模型上保持草图特征;能适应不同用户的绘制习惯,且生成的3维人体模型可用于人体动画设计.结论 提出一种草绘3维人体建模的模板形变方法,支持用户采用草绘方式进行3维人体模型设计,方法具有良好的用户适应性,对3维动画创作具有重要意义.     

基于多尺度HOG的草图检索

草图检索是图像处理领域中的重要研究内容。提出了一种将高斯金字塔和局部HOG特征融合的特征提取改进方法,并将其用于草图检索。采用高斯金字塔将图像分解到多尺度空间,在所有尺度上进行兴趣点提取,获得基于兴趣点的多尺度HOG特征。利用图像的多尺度HOG特征集生成视觉词典,最终形成与视觉词典相关的特征描述向量,通过相似度匹配实现草图检索。将该算法与单一尺度下的HOG算法及其他几种算法比较,实验结果表明了其可行性和有效性。     

Sketch-based garment design with quad meshes

Garment creation continues to be the most tedious part of the virtual clothing process. In this paper, we present an easy to use sketch-based cloth modeling approach. Contours can be easily sketched on a mannequin to generate quad meshes to represent pieces of cloth already fitted and draped. Typically, the clothing process depends greatly on the meshing scheme that has to infer its geometry from the input boundary. Our quad meshing scheme is based on discrete Coons patches but with arbitrary boundary input. We also apply the permanence principle to our topological solution to allow more control over the influence of the input boundary polyline on the interior output polygonal mesh. This facilitates the creation of folds that are strongest in curvature at the boundary and which diminish towards the interior. The generated garments can then be easily animated in a simulation system based on finite elements, using a rediscretization of the generated mesh and a reconstructed metric of the cloth surface.     

结合细粒度特征与深度卷积网络的手绘图检索

目的 传统的手绘图像检索方法主要集中在检索相同类别的图像,忽略了手绘图像的细粒度特征。对此,提出了一种新的结合细粒度特征与深度卷积网络的手绘图像检索方法,既注重通过深度跨域实现整体匹配,也实现细粒度细节匹配。方法 首先构建多通道混合卷积神经网络,对手绘图像和自然图像分别进行不同的处理;其次通过在网络中加入注意力模型来获取细粒度特征;最后将粗细特征融合,进行相似性度量,得到检索结果。结果 在不同的数据库上进行实验,与传统的尺度不变特征（SIFT）、方向梯度直方图（HOG）和深度手绘模型Deep SaN（sketch-a-net）、Deep 3DS（sketch）、Deep TSN（triplet sketch net）等5种基准方法进行比较,选取了Top-1和Top-10,在鞋子数据集上,本文方法Top-1正确率提升了12%,在椅子数据集上,本文方法Top-1正确率提升了11%,Top-10提升了3%,与传统的手绘检索方法相比,本文方法得到了更高的准确率。在实验中,本文方法通过手绘图像能在第1幅检索出绝大多数的目标图像,达到了实例级别手绘检索的目的。结论 提出了一种新的手绘图像检索方法,为手绘图像和自然图像的跨域检索提供了一种新思路,进行实例级别的手绘检索,与原有的方法相比,检索精度得到明显提升,证明了本文方法的可行性。     

Adaptive geometry image

We present a novel post-processing utility called adaptive geometry image (AGIM) for global parameterization techniques that can embed a 3D surface onto a rectangular1 domain. This utility first converts a single rectangular parameterization into many different tessellations of square geometry images(GIMs) and then efficiently packs these GIMs into an image called AGIM. Therefore, undersampled regions of the input parameterization can be up-sampled accordingly until the local reconstruction error bound is met. The connectivity of AGIM can be quickly computed and dynamically changed at rendering time. AGIM does not have T-vertices, and therefore no crack is generated between two neighboring GIMs at different tessellations. Experimental results show that AGIM can achieve significant PSNR gain over the input parameterization, AGIM retains the advantages of the original GIM and reduces the reconstruction error present in the original GIM technique. The AGIM is also for global parameterization techniques based on quadrilateral complexes. Using the approximate sampling rates, the PolyCube-based quadrilateral complexes with AGIM can outperform state-of-the-art multichart GIM technique in terms of PSNR.     

基于跨模态自蒸馏的零样本草图检索

零样本草图检索将未见类的草图作为查询样本,用于检索未见类的图像。因此,这个任务同时面临两个挑战：草图和图像之间的模态差异以及可见类和未见类的不一致性。过去的方法通过将草图和图像投射到一个公共空间来消除模态差异,还通过利用语义嵌入（如词向量和词相似度）来弥合可见类和未见类的语义不一致。在本文中,我们提出了跨模态自蒸馏方法,从知识蒸馏的角度研究可泛化的特征,无需语义嵌入参与训练。具体而言,我们首先通过传统的知识蒸馏将预训练的图像识别网络的知识迁移到学生网络。然后,通过草图和图像的跨模态相关性,跨模态自蒸馏将上述知识间接地迁移到草图模态的识别上,提升草图特征的判别性和泛化性。为了进一步提升知识在草图模态内的集成和传播,我们进一步地提出草图自蒸馏。通过为数据学习辨别性的且泛化的特征,学生网络消除了模态差异和语义不一致性。我们在三个基准数据集,即Sketchy、TU-Berlin和QuickDraw,进行了广泛的实验,证明了我们提出的跨模态自蒸馏方法与当前方法相比较的优越性。     

Optimization methods for truss geometry and topology design

Truss topology design for minimum external work (compliance) can be expressed in a number of equivalent potential or complementary energy problem formulations in terms of member forces, displacements and bar areas. Using duality principles and non-smooth analysis we show how displacements only as well as stresses only formulations can be obtained and discuss the implications these formulations have for the construction and implementation of efficient algorithms for large-scale truss topology design. The analysis covers min-max and weighted average multiple load designs with external as well as self-weight loads and extends to the topology design of reinforcement and the topology design of variable thickness sheets and sandwich plates. On the basis of topology design as an inner problem in a hierarchical procedure, the combined geometry and topology design of truss structures is also considered. Numerical results and illustrative examples are presented.     

Smart image design for large image databases

We propose an active image information system based upon the concept of smart images. A smart image is an image with an associated knowledge structure, consisting of protocols, hot spots, active indexes and attributes. This active image information system enables us to accomplish the objectives of timely delivery and easy accessibility by handling long-duration operations and supporting unitary image information usage. The experimental prototype of the smart image system is described.     

Interrogation of differential geometry properties for design and manufacture

This paper describes a new robust method to decompose a free-form surface into regions with specific range of curvature and provide important tools for surface analysis, tool-path generation, and tool-size selection for numerically controlled machining, tessellation of trimmed patches for surface interrogation and finite-element meshing, and fairing of free-form surfaces. The key element in these techniques is the computation ofall real roots within a finite box of systems of nonlinear equations involving polynomials and square roots of polynomials. The free-form surfaces are bivariate polynomial functions, but the analytical expressions of their principal curvatures involve polynomials and square roots of polynomials. Key components are the reduction of the problems into solutions of systems of polynomial equations of higher dimensionality through the introduction ofauxiliary variables and the use ofrounded interval arithmetic in the context of Bernstein subdivision to enhance the robustness of floating-point implementation. Examples are given that illustrate our techniques.     

Microchannel geometry design for rapid and uniform reagent distribution

Rapid and uniform reagent distribution is critical to the performance of a high-throughput microfluidic system, and its geometric design of the microchannels dominates the accuracy and distribution uniformity of the daughter droplets. This research’s purpose is to optimize the geometry of the T-junction to achieve a uniform distribution of two daughter droplets from a single liquid droplet. Models of gas–liquid flow were realized in the transient numerical simulations to investigate the geometry-dependent pressure distributions and the flowing velocities inside the droplet during the splitting process that leads to an improved design of the T-junction that can increase the stability of the droplet splitting process. To validate that increasing the stability of the splitting process can help improve the distribution uniformity of the daughter droplets, microfluidic devices were manufactured on poly(methyl methacrylate) substrates with micromilling and thermal bonding for experiments. In the multiple experiments, 2 μl of reagent was loaded into the microfluidic device and a uniform pneumatic pressure was applied to push the droplet into the T-junction for splitting. The experimental results, after statistical analysis, show that the improved T-junction can achieve better distribution uniformity of the daughter droplets with a higher reliability and a less reagent loss during the splitting process.     

The fuzzy geometry of image subsets

In pattern recognition we often want to measure geometric properties of regions in an image, but these regions are not always ‘crisply’ defined; it is sometimes more appropriate to regard them as fuzzy subsets of the image. Many of the basic geometric properties of and relationships among regions can be generalized to fuzzy sets; these include connectedness, adjacency and surroundedness, starshapedness and convexity, area and perimeter, extent and diameter. This paper summarizes past work on such fuzzy geometric concepts, and also includes some new results.     

一种投影仪相机系统几何配准的鲁棒算法

目前投影屏的反射及几何特性决定了投影仪投影图像的显示效果。为解除投影仪对标准屏幕的依赖性,借助数码相机可构成反馈系统对显示表面的缺陷进行辐射度补偿和图像校正,而其前提即是建立投影图像面与相机图像面像素之间的配准关系。为实现这一目的提出了一种能实现投影仪相机系统几何配准方法的鲁棒算法。通过设计黑白二色编码图像、显示采集编码图像的方式并对捕获图像进行二值化以及形态学处理以定位和识别采样色块,建立几何配准函数关系。该算法在平面、曲面以及无图案与有图案等各种显示表面的投影场景中得以验证,其平均配准误差稳定在0.2～1.0个像素单位,且不依赖于色块的数量而变化,并能在10s内完成整个配准关系的建立。实验表明该方法能在适用范围以及精度和效率等方面满足相机反馈式投影仪像面配准的应用需求。     

Constrained space deformation techniques for design optimization

We present a novel shape deformation method for its use in design optimization tasks. Our space deformation technique based on moving least squares approximation improves upon existing approaches in crucial aspects: It offers the same level of modeling flexibility as surface-based deformations, but it is independent of the underlying geometry representation and therefore highly robust against defects in the input data. It overcomes the scalability limitations of existing space deformation techniques based on globally supported radial basis functions while providing the same high level of deformation quality. Finally, unlike existing space deformation approaches, our technique directly incorporates geometric constraints–such as preservation of critical feature lines, circular couplings, planar or cylindrical construction parts–into the deformation, thereby fostering the exploration of more favorable and producible shape variations during the design optimization process.     

Floor shape optimization for green building design

Shape is an important consideration in green building design due to its significant impact on energy performance and construction costs. This paper presents a methodology to optimize building shapes in plan using the genetic algorithm. The building footprint is represented by a multi-sided polygon. Different geometrical representations for a polygon are considered and evaluated in terms of their potential problems such as epistasis, which occurs when one gene pair masks or modifies the expression of other gene pairs, and encoding isomorphism, which occurs when chromosomes with different binary strings map to the same solution in the design space. Two alternative representations are compared in terms of their impact on computational effectiveness and efficiency. An optimization model is established considering the shape-related variables and several other envelope-related design variables such as window ratios and overhangs. Life-cycle cost and life-cycle environmental impact are the two objective functions used to evaluate the performance of a green building design. A case study is presented where the shape of a typical floor of an office building defined by a pentagon is optimized with a multi-objective genetic algorithm.