一种具有光照效果和透视效果的图像纹理替换方法

从单幅图像出发,研究图像上具有光照效果和透视效果的图像纹理替换问题。这种方法对于国际互联网上建筑装修时建筑材料的选取有着一定的实际意义。首先提出了一种保持透视效果的图像纹理替换方法,使得替换后的纹理具有与透视相近似的效果,其次讨论了纹理替换后,图像的光亮度和周围环境的统一问题,给出了对替换后纹理进行光亮度调整的方法和实验结果。使用这种方法进行纹理替换的优点在于：替换后的纹理不但保持原有图像的透视效果而且保持原始图像的光照效果,所采用的方法计算量较小、并可实现纹理扩张的效果。实验结果说明该方法是可行的。     

实时图像纹理替换算法

提出了一种实时图像纹理替换算法.使用区域纹理样本实现中等水平视觉的区域纹理替换.利用图像纹理替换潜在的数据并发特征,在集群环境下,设计并实现了混合MPI与OpenMP的图像纹理替换并行算法,给出了主从模式下的MPI并行设计、OpenMP并行设计及负载平衡.实验结果表明,并行算法可行有效,保证重建图像符合人眼视觉要求,大大减少了图像重建所需数据量;有效降低了计算时间,能够满足硬实时系统的时间约束.     

卡通化图像和视频的毛发纹理生成与替换

低成本卡通制作中的图像和视频通常缺乏对动物角色毛发效果的表现,为了能对已有图像及视频中的动物角色进行处理,为其增添具备真实感的毛发效果,提出一种毛发风格化算法——卡通化毛发纹理算法.针对卡通中的动物角色合成毛发纹理并进行替换,分为图像应用及视频应用2个部分.在图像替换时,对要进行风格化处理的目标区域进行图像结构分析,以获取覆盖目标区域的三角网格,再生成毛发纹元并映射于网格之上,通过绘制纹元来生成具备真实感的毛发效果;在进行视频替换时,提取视频关键帧并基于图像应用算法生成相应的卡通化毛发纹理进行图像替换,之后根据关键帧的替换结果指导整个视频的替换.为了获取随时间变化的图像关键帧目标区域,采用SIFT算法计算特征点在时间轴上的匹配;为了快速合成卡通化毛发纹理,采用基于GPU的光线行进算法加速毛发纹元的体绘制过程.实验结果表明,文中算法可成功地对已有图像及视频的动物角色添加具备真实感的毛发效果.     

多区域图像重建算法

为在多区域轮廓曲线中实现中等水平视觉的区域纹理替换,提出了一种多区域图像重建算法。该算法首先提取原始图像的分割区域及其索引号列表;其次根据分段迭代曲线拟合方法获得代表区域结构特征的多项式系数,并选取最小的区域纹理样本;最后利用重建的区域轮廓与合成的区域纹理实现区域恢复。该算法能成功实现包含大量纹理的自然场景图像重建,实验结果表明降低了图像重建所需数据量,适用于图像与视频的静止背景重建。     

抑制纹理信息的偏置场变分图像分割模型

偏置场变分水平集图像分割模型利用原始图像的局部灰度信息,可以对灰度不均匀图像进行有效的分割,但当灰度图像中存在纹理时,分割效果往往很差。针对这一问题,提出抑制纹理信息的偏置场变分水平集图像分割模型。利用一种基于纹理几何结构的纹理描述符描述图像中不同的纹理区域,使得不同纹理区域对比更加明显,相同纹理区域更加平滑,通过抑制纹理信息使后续的图像分割在纹理部分的错分大大减少。实验结果表明,相比偏置场变分模型,所提模型对自然及人工合成纹理图像均获得更好的分割结果。     

基于Wang Tiles的几何纹理合成

提出了一种基于Wang Tiles的几何纹理合成方法来在不同物体表面上即时地生成几何纹理.首先根据给定的几何纹理预计算出一组Wang Tiles,然后用这组Wang Tiles在不同的目标物体上即时生成新的几何纹理.尽管基于Wang Tiles的方法已经应用于图像纹理,但由于几何纹理采用了与图像纹理完全不同的表示方式,因此需要用完全不同的方法来处理.采用了基于约束的几何纹理合成技术自动生成几何纹理Wang Tiles,从而保证了生成的几何纹理Wang Tiles在所有排列下都能保持其几何连续性.与现有的方法相比,生成的几何纹理Wang Tiles可以重用到不同的目标物体上,同时占用的存储空间及计算量更小,速度更快.     

基于单幅图像的真实感人体动画合成

由于单幅图像缺失三维信息以及完整的纹理信息,基于单幅图像的真实感三维人体动画合成极具挑战性。针对单幅图像三维信息缺失问题,提出了一种基于SMPL参数模型的三维人体几何重建方法。该方法以单幅图像为输入,先根据输入图像人体轮廓信息变形标准的SMPL参数模型分别生成与目标轮廓一致的正反面的三维几何模型,然后利用基于B样条插值的网格拼接融合算法拼接正反面三维几何,最后为了恢复正确的手部几何,利用基于B样条插值的网格拼接融合算法,将重建后的模型上错误的手部几何用标准SMPL参数模型上正确的手部几何替换。同时,针对单幅图像中纹理缺失的问题,提出了一个称为FBN（front to back network）的对抗生成网络,用于恢复被遮挡的人体背面纹理。实验结果表明,该方法生成的具有完整纹理的人体几何能够由3D运动数据驱动运动,生成具有高度真实感的三维人体动画。     

基于区间梯度的联合双边滤波图像纹理去除方法

图像纹理去除是指保留图像的语义结构并去除图像中的纹理和噪声部分,从而将纹理区域与结构特征划分开,是计算成像和图像分析的基础研究问题。近年来,在该领域中出现了许多优秀的算法,但它们在结构纹理的区分效果、纹理滤除的干净程度和算法运行的效率等方面仍存在一些问题和矛盾。鉴于此,提出一种基于区间梯度的滤波方法来去除图像纹理。首先,为简化先验模型的复杂度,将区间梯度的概念应用到结构提取模型中,实现了纹理与结构的二元化标记,为下一步单独对纹理区域进行滤波处理做准备。其次,针对复杂多变的纹理部分,将最值替换机制融入到联合双边滤波算法中,即在目标像素的邻域中选取颜色差异最大的像素作为颜色权重分布的中心点,使其在滤波过程中起主导作用。实验证明,所提方法能应用于多种纹理图像,可有效缓解强纹理去除与运算效率之间的矛盾,同时,因在达到相近纹理去除效果时所需迭代次数更少,其在同类滤波方法中取得了更好的边缘保持效果。     

基于非线性滤波和纹理传输的图像风格化

纹理传输通过将样本图像纹理信息传递给源图像实现风格化.但在处理纹理细节丰富的源图像和复杂样本图像时,现有技术的处理效果差,且不便于用户进行直观的交互操作.为此,提出一种基于非线性滤波的显式纹理传输方法.首先通过相对总变差和L0梯度最小化滤波器对源图像进行去纹理操作,然后通过相对总变差滤波器对样本图像进行纹理提取,最后通过亮度重映射和线性叠加实现纹理传输.实验结果表明,该方法能够在保持源图像结构和色调信息的同时进行纹理替换;与已有方法相比,交互过程更加直观且风格化效果更好.     

基于人脸标准化的纹理和光照保持3D人脸重构

现有人脸纹理重建方法对于人脸的皱纹、胡须、瞳孔颜色等重建效果往往不够细致.为了解决此问题,文中提出基于人脸标准化的纹理和光照保持3D人脸重构.首先对2D人脸图像标准化,使用光照信息和对称纹理重构人脸自遮挡区域的纹理.然后依据2D-3D点对应关系从标准化的2D人脸图像获取相应的3D人脸纹理,结合人脸形状重构和纹理信息,得到最终的3D人脸重构结果.实验表明文中方法有效保留原始2D图像的纹理和光照信息,重构的人脸更自然,具有更丰富的人脸细节.     

Mesh-guided optimized retexturing for image and video

This paper presents an approach of replacing textures of specified regions in the input image and video using stretch-based mesh optimization.. The retexturing results have the similar distortion and shading effects conforming to the underlying geometry and lighting conditions. For replacing textures in single image,two important steps are developed: the stretch-based mesh parametrization incorporating the recovered normal information is deduced to imitate perspective distortion of the region of interest; the Poisson-based refinement process is exploited to account for texture distortion at fine scale.The luminance of the input image is preserved through color transfer in YCbCr color space. Our approach is independent of the replaced textures. Once the input image is processed, any new texture can be applied to efficiently generate the retexturing results. For video retexturing, we propose key-frame-based texture replacement extended and generalized from the image retexturing. Our approach repeatedly propagates the replacement result of key frame to the rest of the frames. We develop the local motion optimization scheme to deal with the inaccuracies and errors of robust optical flow when tracking moving objects. Visibility shifting and texture drifting are effectively alleviated using graphcut segmentation algorithm and the global optimization to smooth trajectories of the tracked points over temporal domain. Our experimental results showed that the proposed approach can generate visually pleasing results for both image and video.     

Synthesis of Color Textures for Multimedia Applications

An algorithm for synthesizing color textures from a small set of parameters is presented in this paper. The synthesis algorithm is based on the 2-D moving average model, and realistic textures resembling many real textures can be synthesized using this algorithm. A maximum likelihood estimation algorithm to estimate parameters from a sample texture is also presented. By combining the estimation and synthesis algorithms, a color texture can be synthesized from a sample texture without human intervention. Using the estimated parameters, a texture larger than the original image can be synthesized from a small texture sample. The synthesis algorithm does not require an expensive iterative algorithm, and the quality of synthesized textures may be acceptable for many multimedia applications. In the experiment, various textures suitable for multimedia applications are synthesized from parameters estimated from real textures.     

Fast and accurate texture placement

Mapping from parameter space to texture space separates textures from objects, enabling control of textured images without modifying the object or texture. Although conceptually straightforward, γ-mapping turns out to be very useful in practical texture mapping implementation. It logically separates the texture from the object and makes the texture mapping process flexible and easier to use. By modifying the γ-mapping, texture placement on the object is efficiently controlled without modifying either the texture image or the underlying geometry. We concentrate on efficiency issues related to γ-mapping, as well as its practical applications to parametric surfaces     

Capture and fusion of 3d surface texture

Image fusion is a process that multiple images of a scene are combined to form a single image. The aim of image fusion is to preserve the full content and retain important features of each original image. In this paper, we propose a novel approach based on wavelet transform to capture and fusion of real-world rough surface textures, which are commonly used in multimedia applications and referred to as3D surface texture. These textures are different from 2D textures as their appearances can vary dramatically with different illumination conditions due to complex surface geometry and reflectance properties. In our approach, we first extract gradient/height and albedo maps from sample 3D surface texture images as their representation. Then we measure saliency of wavelet coefficients of these 3D surface texture representations. The saliency values reflect the meaningful content of the wavelet coefficients and are consistent with human visual perception. Finally we fuse the gradient/height and albedo maps based on the measured saliency values. This novel scheme aims to preserve the original texture patterns together with geometry and reflectance characteristics from input images. Experimental results show that the proposed approach can not only capture and fuse 3D surface texture under arbitrary illumination directions, but also has the ability to retain the surface geometry properties and preserve perceptual features in the original images.     

多种子快速纹理合成

为避免纹理映射引起的接缝和扭曲,近几年人们提出了一些纹理合成方法。但大多数只能处理部分纹理,而且相当费时,为此,提出一种新的纹理合成算法,该算法在样图中选取多个种子,并将其预置到合成图中作为实始匹配点,然后通过螺旋状路径,在初始匹配点领域搜索寻找新匹配点进行合成,以此循环,直至合成图充满为止,该方法大大加快了合成速度,与穷尽搜索法相比,合成速度平均可以提高360倍,特别是在亮度空间合成时,在合成图象质量与RGB空间结果基本相同的情况下,可以实现进一步加速,另外,该算法还可合成沿方向变化的纹理,对不同纹理进行合成实验,其结果令人满意。     

Solid texture synthesis for heterogeneous translucent materials

We present a method to synthesize solid textures from heterogeneous translucent materials that have a complex pattern and subsurface scattering effect. A solid texture provides consistent texture throughout the volume, so that it can be used to model the texture on an arbitrary geometry. However, solid texture synthesis requires a huge amount of time to generate the volume. Moreover, a synthesized solid texture acquires only the color information from an input exemplar. Therefore, it has been difficult to render the appearance of a translucent object realistically without additional appearance data. In this paper, we introduce a new search method to accelerate synthesizing of solid textures. This method decomposes the candidates in an exemplar into several subgroups and searches for the best similar neighborhood in each decomposed subgroup. We also apply subsurface scattering effects to the shell layer of a synthesized object for realistic rendering of a translucent solid texture. Experimental results show that our rendering method can produce realistic rendering results for various heterogeneous translucent objects. It can also represent cross-sections of an object realistically without reconstructing the texture and surface geometry.     

Projective texture atlas construction for 3D photography

The use of attribute maps for 3D surfaces is an important issue in geometric modeling, visualization and simulation. Attribute maps describe various properties of a surface that are necessary in applications. In the case of visual properties, such as color, they are also called texture maps. Usually, the attribute representation exploits a parametrization g:U??²→?³ of a surface in order to establish a two-dimensional domain where attributes are defined. However, it is not possible, in general, to find a global parametrization without introducing distortions into the mapping. For this reason, an atlas structure is often employed. The atlas is a set of charts defined by a piecewise parametrization of a surface, which allows local mappings with small distortion. Texture atlas generation can be naturally posed as an optimization problem where the goal is to minimize both the number of charts and the distortion of each mapping. Additionally, specific applications can impose other restrictions, such as the type of mapping. An example is 3D photography, where the texture comes from images of the object captured by a camera [4]. Consequently, the underlying parametrization is a projective mapping. In this work, we investigate the problem of building and manipulating texture atlases for 3D photography applications. We adopt a variational approach to construct an atlas structure with the desired properties. For this purpose, we have extended the method of Cohen–Steiner et al. [6] to handle the texture mapping set-up by minimizing distortion error when creating local charts. We also introduce a new metric tailored to projective maps that is suited to 3D photography.     

基于分块的高效图像可逆认证方法

图像可逆认证是一项将可逆信息隐藏和脆弱水印相结合的新技术,其既能实现对图像的脆弱认证,还能在提取认证信息的同时无失真地恢复出原始载体,对图像的原始性和完整性认证具有非常重要的意义.针对现有可逆认证方法认证精度低、对具有复杂纹理的图像或图像中部分纹理复杂区域无法实现有效保护的问题,提出一种新的图像可逆认证方法.首先对待认证图像进行分块,根据每个子块可嵌入容量将其分为差分块和平移块,并采用不同的可逆嵌入方法对不同类型的块进行认证码嵌入操作.为了增大嵌入容量以提高对每个子块的认证效果,还采取了分层嵌入的方式.在认证方,可以通过从每个子块中提取认证码实现子块的篡改检测和定位.此外,所提方法还可与形态学中的膨胀和腐蚀操作结合以细化篡改检测标记,进一步提高检测效果.实验结果表明,所提方法能够在同样的认证精度下对纹理平滑和纹理复杂的图像进行保护,同时还能够实现对几乎所有子块的独立认证和恢复,具有广泛的适用性.     

高光去除的聚类算法改进

高光去除是计算机视觉领域研究的一个热点问题.现有的基于双色反色模型分离漫反射分量和镜面反射分量去除单幅图像中的高光的方法,容易引起图像颜色失真和纹理的丢失.针对此问题,在使用像素强度比去高光的基础上改进了像素聚类算法,能够更准确的进行像素分类,改善图像颜色失真的现象.首先计算原图像与最小强度值单通道图像的差值得到无高光图像.然后根据无高光图像计算与高光区域相关的每个像素点的最大漫反射色度值和最小漫反射色度值.最后将高光区域内的像素点转换到最小最大色度空间,对高光区域内的像素点进行xmeans聚类,利用分类后漫反射像素点的强度比估计值很容易分离高光区域像素点的镜面反射分量,从而得到去高光图像.实验结果表明,与现有的方法对比,峰值信噪比值平均提升了2%至4%,图像颜色失真和纹理丢失状况得到改善,视觉效果更好.     

基于几何与图像混合绘制中的快速WARP变换算法研究

基于几何与图像的混合绘制中，3D Warp算法以严格的数学变换为基础，从而能够保证准确的投影关系，但该算法在实时绘制阶段需进行大量的数学运算，故其时间复杂度较高，该文提出了一种新的快速Warp变换算法，算法以3D Warp算法为基础，采用了崭新的投影过程，从而使时间复杂度较3D Warp算法有较大幅度的下降（降低约7．51倍）。同时，该算法是一种流水结构，能够有效利用现有的加速硬件，而无需改变图形硬件的体系结构。