Exemplar‐Based Portrait Photograph Enhancement as Informed by Portrait Paintings

This paper proposes an approach to enhance the regional contrasts in snap‐shot style portrait photographs by using pre‐modern portrait paintings as aesthetic exemplars. The example portrait painting is selected based on a comparison of the existing contrast properties of the painting with those of the photograph. The contrast organization in the selected example painting is transferred to the photograph by mapping the inter‐ and intra‐regional contrasts of the regions, such as the face and skin areas of the foreground figure, the non‐face/skin part of the foreground and the background region. A piecewise non‐linear transformation curve is used to achieve the contrast mapping. Finally, the transition boundary between regions is smoothed to achieve the final results. The experimental results and user study demonstrate that, by using this proposed approach, the visual appeal of the portrait photographs is effectively improved, and the face and the figure become more salient.     

基于信息熵的风格绘画分类研究

针对艺术风格绘画分类算法中存在的精度和效率不高等问题,提出一种基于信息 熵的艺术风格绘画分类算法。首先选取西方漫画、素描、油画、水彩画,以及国内烙画、水墨 画、壁画具有代表性的 7 种艺术绘画风格作为研究对象,对图像进行去噪、归一化等预处理。 其次,提取绘画艺术作品风格特征,分别求取图像的颜色熵、分块熵、轮廓熵,并合并构成不 同绘画风格的信息熵。信息熵求取时,将色彩空间转换为 Lab 颜色空间,通过 a、b 通道颜色值 及加权函数获得图像的颜色熵;通过对艺术图像分块求取信息熵,求取分块的信息熵均值获得 分块熵;通过 Contourlet 变换,求取艺术图像的轮廓信息,获得轮廓熵。接着,合并提取的颜 色熵、分块熵、轮廓熵,利用支持向量机(SVM)对艺术风格图像学习训练,获得艺术绘画风格 的分类模型;最后,提取待识别绘画风格样本的熵特征,通过 SVM 分类识别获得最终的分类 结果。该方法具有特征维数少、运算速度快、尺度不变性等优势,实验结果表明,其能提高不 同绘画风格的分类精度和效率。     

Fast generation method of 3D scene in Chinese landscape painting

The three-dimensional (3D) modeling of Chinese landscape painting is of great significance for the digital protection of cultural heritage and the production of virtual reality content. A fast modeling method to create 3D landscape scenes for traditional Chinese painting is proposed in this paper, based on integrated terrain modeling and the water flow rendering algorithm. A height map generation algorithm based on auxiliary lines is first proposed to carry out fast modeling from a simple two-dimensional contour to create a 3D mountain model. A realistic flow simulation that fits the topography is then undertaken, based on a flow chart which is calculated using the particle force in the normal grid of topography, and the theory of smoothing particle hydrodynamics. Finally, a stylistic scene that conforms to the artistic concept of traditional Chinese painting is acquired by optimizing the parameters. The interactive modeling platform of the integrated algorithm is tested in this study, and compared with existing research. Results show the method can achieve real-time rendering and realistic rendering to rapidly generate a 3D scene model consistent with a traditional painting scene, and provide support for the follow up development of virtual reality applications.

     

Robust radiometric calibration and vignetting correction

In many computer vision systems, it is assumed that the image brightness of a point directly reflects the scene radiance of the point. However, the assumption does not hold in most cases due to nonlinear camera response function, exposure changes, and vignetting. The effects of these factors are most visible in image mosaics and textures of 3D models where colors look inconsistent and notable boundaries exist. In this paper, we propose a full radiometric calibration algorithm that includes robust estimation of the radiometric response function, exposures, and vignetting. By decoupling the effect of vignetting from the response function estimation, we approach each process in a manner that is robust to noise and outliers. We verify our algorithm with both synthetic and real data which shows significant improvement compared to existing methods. We apply our estimation results to radiometrically align images for seamless mosaics and 3D model textures. We also use our method to create high dynamic range (HDR) mosaics which are more representative of the scene than normal mosaics.     

Single-Image Vignetting Correction

In this paper, we propose a method for robustly determining the vignetting function given only a single image. Our method is designed to handle both textured and untextured regions in order to maximize the use of available information. To extract vignetting information from an image, we present adaptations of segmentation techniques that locate image regions with reliable data for vignetting estimation. Within each image region, our method capitalizes on the frequency characteristics and physical properties of vignetting to distinguish it from other sources of intensity variation. Rejection of outlier pixels is applied to improve the robustness of vignetting estimation. Comprehensive experiments demonstrate the effectiveness of this technique on a broad range of images with both simulated and natural vignetting effects. Causes of failures using the proposed algorithm are also analyzed.     

Boosting image classification with LDA-based feature combination for digital photograph management

Image classification is of great importance for digital photograph management. In this paper we propose a general statistical learning method based on boosting algorithm to perform image classification for photograph annotation and management. The proposed method employs both features extracted from image content (i.e., color moment and edge direction histogram) and features from the EXIF metadata recorded by digital cameras. To fully utilize potential feature correlations and improve the classification accuracy, feature combination is needed. We incorporate linear discriminant analysis (LDA) algorithm to implement linear combinations between selected features and generate new combined features. The combined features are used along with the original features in boosting algorithm for improving classification performance. To make the proposed learning algorithm more efficient, we present two heuristics for selective feature combinations, which can significantly reduce training computation without losing performance. The proposed image classification method has several advantages: small model size, computational efficiency and improved classification performance based on LDA feature combination.     

Evaluation of texture registration by epipolar geometry

In the process of digitizing the geometry and appearance of 3D objects, texture registration is a necessary step that solves the 2D–3D mapping between the 2D texture images and the 3D geometric model. For evaluation of texture registration with ground truth, accurate datasets can be obtained with a complex setup consisting of calibrated geometry and texture capture devices. We do not have any knowledge of such evaluation performed; current evaluations reflect, at their best, the precision achieved by the algorithms, but fail to identify a possible bias. We propose a new evaluation measure based on the epipolar geometry of texture image pairs, with the advantage that the ground truth can be extracted solely from the texture images, independent of the 3D acquisition. We developed a noise model suitable to our purpose and analysed three distance measures based on epipolar geometry, well known in the computer vision community, to find new theoretical and experimental results. Finally, using the proposed framework, we evaluated a texture registration algorithm based on mutual information and found that its accuracy was under half-pixel.     

图像驱动的三维模型外观细节增强

随着计算机游戏与电影视觉特效的普及应用,仅通过纹理贴图已不能满足用户需求。然而,为三维模型制作视觉真实的表面外观,即在包含几何细节变化(如金属腐蚀、石料风化、木质裂纹等)同时辅之以与几何变化合理匹配的材质颜色却是一件困难且耗时的工作。文章提出了一种能够从单幅图像中提取物体的外观细节(材质与几何信息)并增强至三维模型表面的方法。该方法是一种新型的几何约束的非刚性2D-3D融合配准算法,可以通过将配准后的模型提供的几何信息作为先验知识实现更好的本征图像分解结果。利用图像中同步分解出的互相关的多尺度几何信息与材质纹理信息组成的非参数外观模型,文中提出了一种法向量约束的几何变形算法将外观纹理恢复到代理模型上。通过上述步骤,文章提出的方法能够帮助建模师制作具有多尺度外观细节的三维模型。     

An Adaptive Gaussian Mixture Model for Non-rigid Image Registration

This paper presents a new model based on statistical and variational methods for non-rigid image registration. It can be viewed as an improvement of the intensity-based model whose dissimilarity term is based on minimization of the so-called sum of squared difference(SSD). In the proposed model, it is assumed that the residue of two images can be described as a mixture of Gaussian distributions. Then we incorporate the features of variational regularization methods and expectation-maximization(EM) algorithm, and propose the new model. The novelty is the introduction of two weighting functions and some control parameters in dissimilarity term. The weighting functions could identify low and high contrast objects of the residue automatically and effectively, and the control parameters help to improve the robustness of the model to the choice of regularization parameters. By the introduced parameters and weighting functions, the algorithm could locally adjust the behavior of deformation in different contrast regions. Numerical experimental results of 2D synthetic and 3D MR brain images demonstrate the efficiency and accuracy of the proposed approach compared with other methods.     

国画水墨扩散的LBE计算机模拟

为了满足人们在美工设计中对计算机国画艺术效果仿真的需求,研究和分析了水墨扩散LBE(lattice Boltzmann equation,网格波尔兹曼)算法,提出运用亮度决定水粒子的初始分布,用CMY粒子进行扩散模拟.通过对LBE算法进一步推导而提出疑问,并对扩散及受阻方程进行修正.使用Wintab 1.1压感设备接口完成用户压感数据的读取,借助图像处理函数库IPL98 2.0进行扩散实验数据结构和实验流程设计,实现了国画水墨扩散效果的模拟,证明了该方法的可行性和有效性.     

组合式建筑立面全景图像的渐晕校正

由于组合式建筑立面图像进行全景拼接时,交界处光照度降低,产生渐晕现象,提出一种基于曲面拟合的渐晕校正方法,对组合式建筑立面全景图像渐晕校正.通过推算角点,分析图像坐标系及世界坐标系之间的对照关联,获取世界坐标系中角点的单位坐标数据,构建相机成像的组合式建筑三维交互模型;采用ORB特征匹配算法实施图像匹配,通过分解单应性...     

一种基于曲率与置信度传播的视差估计方法

常用的匹配代价容易受到光晕效应等复杂光学畸变的影响。文中通过对相机成像模型的分析,从理论上证明基于曲率的匹配代价对光晕具有较强的鲁棒性。提出一种在积分图像上对基于曲率的匹配代价进行计算的方法,有效提高算法速度。最后,将曲率约束引入到置信度传播算法正则项的设计中,抑制置信度在深度不连续区域之间的传递,防止过平滑现象的产生。实验结果验证文中方法的有效性。     

基于几何特征的人脸定位算法的改进

针对鼻子在照片中往往因与脸部皮肤的对比太不强烈而根本不出现在二值化图像中这一问题,在原有几何匹配算法的基础上给出了鼻子的评价函数,得到了鼻子标记的定位算法,从而提高了人脸定位的精度。     

基于属性分辨度的最大相容块规则提取算法

提出一种基于属性分辨度的不完备决策表规则提取算法, 它是一种例化方向的方法. 首先从空集开始, 逐步 选择当前最重要的条件属性对对象集分类, 从广义决策值唯一的相容块提取确定规则, 从其他的相容块提取不确定 规则; 然后设计属性必要性判断步骤去除每条规则的冗余属性; 最后通过规则约简过程来简化所获得的规则, 增强规 则的泛化能力. 实验结果表明, 所提出的算法效率更高, 并且所获得的规则简洁有效.

     

面向自然图像的年画风格化算法

年画作为中国民间艺术的优秀代表之一,其风格对艺术创作有重要意义。为了快速方便地实现自然图像的年画风格化,设计了一种自动化的年画风格化算法框架,并尽可能准确的保持自然图像的原有图案与颜色分配。对于一幅用户给定的自然图像,算法首先对图像进行自动的颜色分割聚类,提取关键线条作为年画轮廓线,运用基于优化的年画配色映射方法将不同的分割区域重新着色,最后通过年画风格细节增强,使结果图的颜色风格更加接近年画风格。针对颜色配色优化,算法综合考量了生成图像与原图像的颜色相似度和相邻色块的颜色对比度,并利用差分进化策略对能量方程进行最小化求解得出合理的颜色配置方案。在实验中进行了几组不同产地的年画风格化实验,结果算法可有效地将一幅自然图像自动转化为年画风格,且效果较自然,颜色具有地方特色。     

基于颜色特征的画家艺术风格提取方法

针对现有全局与局部特征提取方法所提取的颜色特征无法有效描述画家艺术风格的问题，提出了一种基于关键区域的油画描述法。首先，通过融合主色调占比与颜色丰富度定义了油画区域信息丰富度，以检测并选取油画的关键区域。其次，提取所选油画关键区域的54维特征来描述油画，并利用Fisher Score对这些特征进行评估，选取重要的特征描述油画关键区域，从而高度体现画家艺术风格。最后，为了验证上述方法的有效性，利用朴素贝叶斯分类器实现油画的分类。建立两个数据库进行仿真实验，数据库1包含3位画家的120幅作品，数据库2包含9位画家3种流派的513幅作品。数据库1上的实验结果表明，结合Fisher Score特征描述的分类正确率高于普通分类的正确率，所提方法依据画家与流派的油画分类正确率分别达到了90%与90.20%。数据库2上的实验结果表明，所提方法根据画家的油画分类正确率达到了90%，相较Features-FS的正确率提高了6.67个百分点；依据流派分类的结果正确率达到了90.20%，与Features-FS的正确率相当。所提的基于关键区域的油画描述法所提取的特征能够有效描述画家的艺术风格。     

Topology based 2D engineering drawing and 3D model matching for process plant

Process plant models mainly include 3D models and 2D engineering drawings. Matching calculation between these CAD models has wide applicability in model consistency check and retrieval. In process plant, engineering design standards make 2D engineering drawing and 3D model differ in geometry, proportion and structure, leading to the inapplicability of current shape-feature based matching approaches. Since connection relationships between components are the core of a process plant, a topology based algorithm is proposed. Firstly, by exploiting components as vertices and relationships as edges, both 2D engineering drawing and 3D model are preprocessed into graph structures. Then each model’s relationship types are extracted from the graph. Finally, regarding the extracted relationship types as primary feature, feature similarity is calculated to measure the matching degree between their corresponding models. The proposed algorithm is geometric deformation invariant. Experiments with industrial applications are presented, which demonstrates the effectiveness and feasibility of the proposed algorithm.     

基于模拟退火算法的波束图设计

文中提出了基于模拟退火算法的波束设计方法.其思想是按照给定的波束优化要求,建立某种相应的目标函数,利用模拟退火算法对波束的加权系数进行全局寻优,达到波束优化的目的.模拟退火算法模拟了固体缓慢退火从而获得最低内能的物理过程,是一种简单而有效的随机性全局优化算法.为证明该设计方法的可行性和有效性,给出了15元平面阵的设计实例以及仿真结果.结果表明,使用该方法设计的波束与常规方法相比,获得了更低的旁瓣级.     

Multiplicative approximation of transfer functions with frequency weighting

An algorithm for transfer function order reduction is presented, which generalizes the balanced stochastic truncation algorithm to allow for input weighting. An example illustrates use of the algorithm to secure smaller dB error in selected frequency bands through the introduction of the weighting.     

Virtual viewpoints reconstruction via Fourier slice transformation

We propose a new method of calculating the arbitrary viewpoints for auto‐stereoscopic display. The three‐dimensional (3D) object is first virtually reconstructed in 3D spaces by mapping each pixel with a depth according to the depth mapping. We then calculate the Fourier spectrum of the 3D object by the fast Fourier transformation. The arbitrary viewpoints are reconstructed by “slicing” the 3D Fourier spectrum. To repair “black hole” artifacts, the regions in the background are calculated by advanced boundary in‐painting. Experimental results show the effectiveness and accuracy of the proposed algorithm to calculate viewpoints with arbitrary viewing angles. A comparison is also presented, which indicates that the proposed algorithm is more accurate than conventional method, and the advanced boundary in‐painting can save three quarters of time than the conventional in‐painting method.