Polyhedral Objects Metamorphosis Using Convex Decomposition and Morphology

A new technique is presented for computing continuous shape transformations between polyhedral objects. The polyhedron shape transformations can be divided into polyhedron metamorphosis and bi-directional local rigid body rotation transformation. By decomposing two objects into sets of individual convex sub-objects respectively, and establishing the matching between two subsets, the approach can solve the metamorphosis problem of two non-homotopic objects (including concave objects and holey objects). Compared with other methods, this metamorphosis algorithm can be executed automatically for arbitrary polyhedrons and no need user interaction. The user has the ability to choose an automatic matching or to select interactively pairs of corresponding matching convex subsets to obtain special effects. Experiments show that this method can generate natural, high-fidelity, eye-pleasing metamorphosis results with simple computation.     

自然雪景的构造和绘制

合成高度复杂自然场景一直是计算机图形学中的一个富有挑战性的课题。生成雪景或许更加困难，因为利用常用的造型工具，比如多边形和曲面来表达雪的形状非常不方便。该文介绍一种解决这个问题的方法，这种方法采用位移映射（displacement mapping）来构造近处由多边形构造的物体上的雪块，而体纹理映射（volumetric texture mapping）被用来处理远处的复杂物体，比如树木。绘制结果显示，这种方法能够在较少的内存和计算开销的基础上绘制出高度复杂和真实的雪景。     

Visual learning and recognition of 3-d objects from appearance

The problem of automatically learning object models for recognition and pose estimation is addressed. In contrast to the traditional approach, the recognition problem is formulated as one of matching appearance rather than shape. The appearance of an object in a two-dimensional image depends on its shape, reflectance properties, pose in the scene, and the illumination conditions. While shape and reflectance are intrinsic properties and constant for a rigid object, pose and illumination vary from scene to scene. A compact representation of object appearance is proposed that is parametrized by pose and illumination. For each object of interest, a large set of images is obtained by automatically varying pose and illumination. This image set is compressed to obtain a low-dimensional subspace, called the eigenspace, in which the object is represented as a manifold. Given an unknown input image, the recognition system projects the image to eigenspace. The object is recognized based on the manifold it lies on. The exact position of the projection on the manifold determines the object's pose in the image.A variety of experiments are conducted using objects with complex appearance characteristics. The performance of the recognition and pose estimation algorithms is studied using over a thousand input images of sample objects. Sensitivity of recognition to the number of eigenspace dimensions and the number of learning samples is analyzed. For the objects used, appearance representation in eigenspaces with less than 20 dimensions produces accurate recognition results with an average pose estimation error of about 1.0 degree. A near real-time recognition system with 20 complex objects in the database has been developed. The paper is concluded with a discussion on various issues related to the proposed learning and recognition methodology.     

Novel View Synthesis Of Transparent Object From a Single Image

We propose a method for converting a single image of a transparent object into multi-view photo that enables users observing the object from multiple new angles, without inputting any 3D shape. The complex light paths formed by refraction and reflection makes it challenging to compute the lighting effects of transparent objects from a new angle. We construct an encoder–decoder network for normal reconstruction and texture extraction, which enables synthesizing novel views of transparent object from a set of new views and new environment maps using only one RGB image. By simultaneously considering the optical transmission and perspective variation, our network learns the characteristics of optical transmission and the change of perspective as guidance to the conversion from RGB colours to surface normals. A texture extraction subnetwork is proposed to alleviate the contour loss phenomenon during normal map generation. We test our method using 3D objects within and without our training data, including real 3D objects that exists in our lab, and completely new environment maps that we take using our phones. The results show that our method performs better on view synthesis of transparent objects in complex scenes using only a single-view image.     

A New Method of Image Mosaicking and Its Application to Cultural Heritage Representation

This paper presents an original two-step procedure for estimating projective transformations between pairs of images: first, the transformation between the images is approximated as an affine transformation; second, this estimate is refined into that of a projective transformation. This strategy for matching projective views is computationally very efficient. The proposed method can be applied both to mosaicking of high resolution images of planar textured objects (e.g., frescoes and paintings), with subpixel accuracy, and to construction of panoramic images. Practical examples of mosaicking of cultural heritage imagery obtained by using the presented procedure are discussed in the paper.     

Image Warping for Shape Recovery and Recognition

We demonstrate that, for a large class of reflectance functions, there is a direct relationship between image warps and the corresponding geometric deformations of the underlying three-dimensional objects. This helps explain the hidden geometrical assumptions in object recognition schemes which involve two-dimensional image warping computed by matching image intensity. In addition, it allows us to propose a novel variant of shape from shading which we call shape from image warping. The idea is that the three-dimensional shape of an object is estimated by determining how much the image of the object is warped with respect to the image of a known prototype shape. Therefore, detecting the image warp relative to a prototype of known shape allows us to reconstruct the shape of the imaged object. We derive properties of these shape warps and illustrate the results by recovering the shapes of faces.     

Describing a Robot's Workspace Using a Sequence of Views from a Moving Camera

This correspondence describes a method of building and maintaining a spatial respresentation for the workspace of a robot, using a sensor that moves about in the world. From the known camera position at which an image is obtained, and two-dimensional silhouettes of the image, a series of cones is projected to describe the possible positions of the objects in the space. When an object is seen from several viewpoints, the intersections of the cones constrain the position and size of the object. After several views have been processed, the representation of the object begins to resemble its true shape. At all times, the spatial representation contains the best guess at the true situation in the world with uncertainties in position and shape explicitly represented. An octree is used as the data structure for the representation. It not only provides a relatively compact representation, but also allows fast access to information and enables large parts of the workspace to be ignored. The purpose of constructing this representation is not so much to recognize objects as to describe the volumes in the workspace that are occupied and those that are empty. This enables trajectory planning to be carried out, and also provides a means of spatially indexing objects without needing to represent the objects at an extremely fine resolution. The spatial representation is one part of a more complex representation of the workspace used by the sensory system of a robot manipulator in understanding its environment.     

Virtual Dunhuang Art Cave: A Cave within a CAVE

Virtual Reality can present historical places in a three-dimensional and interactive way, giving visitors a photorealistic impression of objects. Not only existing scenarios can be shown, but VR can also be used to rebuild scenarios that were damaged or destroyed a long time ago, giving new life to the cultural heritage. We used Virtual Reality to present the Mogao Grottos in Dunhuang. This cave site is one of the most important cultural and religious places by the ancient Silk Road. The presentation is to give visitors the impression of visiting the cave site and provide information about the caves, paintings and statues in an interesting way. To achieve this, we developed a new, intuitive interaction paradigm, which enables the user to explore the caves. To give observers a photorealistic impression of the caves and to create a feeling of immersion, innovative rendering techniques were integrated.
The resulting presentation combines Virtual Reality and archaeology to give tourists a realistic experience of this cave site and to support scientists in their research work.     

利用形体特征的铅笔素描画生成

铅笔素描是以单色线条来表现物体的艺术。画家观察物体的形体结构,以线塑型表现物体明暗层次和形体特征。为了表现物体的立体感,提出一种利用形体特征的铅笔素描画生成方法。首先对图像进行双边滤波处理,去除图像微小细节;然后由亮度图像计算其向量场,根据物体的形体特征对向量进行平滑,使其符合原图像的形体特征;最后利用线积分卷积的方法生成铅笔素描纹理,调整明暗对比后得到铅笔素描的效果。实验结果表明,该方法能有效地绘制铅笔素描的立体感和明暗关系。     

Graphical objects

We introduce the concept of graphical objects, the abstraction paradigms to study them, and their applications in computer graphics. Intuitively, graphical objects encompass all the entities manipulated in a graphics system. This notion makes it possible to unify similar research topics appearing in the literature separately. We study the problem of object metamorphosis, which includes the problem of image metamorphosis. Although we are not primarily concerned with implementation issues in this paper, the concepts we introduce can be exploited for system design and development that use object-oriented programming.     

一种二维极坐标傅里叶描述子在图像检索中的应用

在图像检索中,形状描述是图像内容描述的关键部分。目前已经存在的大多数形状描述符均是基于某一特定应用的,不具有通用性。文章提出了一种基于二维极坐标傅里叶变换的傅里叶描述子对目标形状的描述方法,分析并实验验证了该方法的各个性质,并讨论了它的实际计算问题。实验结果表明,该描述符通用性强,鲁棒性高。     

CT序列图像分割的实现及分割结果的重建

主要采用模糊边缘增强算法与B样条蛇线算法结合的方法实现CT图象中某些组织器官二维上的分割，并采用体绘制对分割后的序列图像进行三维重建。图像二维切片和三维重建的结果表明该方法的效果良好。程序作为基于Java的医学图像分析系统的子功能使用。     

A Translucent Sketchpad for the Virtual Table Exploring Motion-based Gesture Recognition

The Virtual Table presents stereoscopic graphics to a user in a workbench-like setting. For this device, a user interface and new interaction techniques have been developed based on transparent props -a tracked hand-held pen and a pad. These props, particularly the pad, are augmented with 3D graphics from the Virtual Table's display that can serve as a palette for tools and controls as well as a window-like see-through interface, a plane-shaped and through-the-plane tool, supporting a variety of new interaction techniques. This paper reports on an extension of this user-interface design space which uses gestural input to create and control solid geometries for CAD and conceptual design. The application of gestural interfaces is a common method for interacting with virtual environments on a habitual and natural basis. The motion-based gesture recognition presented here uses Fuzzy Logic to support a predictable, flexible, and efficient learning process. This new interaction paradigm greatly increases the Virtual Table's suitability for design tasks. Traditional CAD dialogue can be combined with intuitive rapid sketching of geometry on the pad. Additionally, the resulting events and objects can be associated with scene details below the translucent tablet.     

Texturing 3D Models of Real World Objects from Multiple Unregistered Photographic Views

As the efficiency of computer graphic rendering methods is increasing, generating realistic models is now becoming a limiting factor. In this paper we present a new technique to enhance already existing geometry models of real world objects with textures reconstructed from a sparse set of unregistered still photographs. The aim of the proposed technique is the generation of nearly photo-realistic models of arbitrarily shaped objects with minimal effort. In our approach, we require neither a prior calibration of the camera nor a high precision of the user's interaction. Two main problems have to be addressed of which the first is the recovery of the unknown positions and parameters of the camera. An initial estimate of the orientation is calculated from interactively selected point correspondences. Subsequently, the unknown parameters are accurately calculated by minimising a blend of objective functions in a 3D-2D projective registration approach. The key point of the proposed method of registration is a novel filtering approach which utilises the spatial information provided by the geometry model. Second, the individual images have to be combined yielding a set of consistent texture maps. We present a robust method to recover the texture from the photographs thereby preserving high spatial frequencies and eliminating artifacts, particularly specular highlights. Parts of the object not seen in any of the photographs are interpolated in the textured model. Results are shown for three complex example objects with different materials and numerous self-occlusions.     

A Practical Analysis of Clustering Strategies for Hierarchical Radiosity

The calculation of radiant energy balance in complex scenes has been made possible by hierarchical radiosity methods based on clustering mechanisms. Although clustering offers an elegant theoretical solution by reducing the asymptotic complexity of the algorithm, its practical use raises many difficulties, and may result in image artifacts or unexpected behavior. This paper proposes a detailed analysis of the expectations placed on clustering and compares the relative merits of existing, as well as newly introduced, clustering algorithms. This comparison starts from the precise definition of various clustering strategies based on a taxonomy of data structures and construction algorithms, and proceeds to an experimental study of the clustering behavior for real-world scenes. Interestingly, we observe that for some scenes light is difficult to simulate even with clustering. Our results lead to a series of observations characterizing the adequacy of clustering methods for meeting such diverse goals as progressive solution improvement, efficient ray casting acceleration, and faithful representation of object density for approximate visibility calculations.     

ASSET-2: Real-Time Motion Segmentation and Object Tracking

This paper describes how image sequences taken by a moving video camera may be processed to detect and track moving objects against a moving background in real-time. The motion segmentation and shape tracking system is known as a scene segmenter establishing tracking, version 2 (ASSET-2). Motion is found by tracking image features, and segmentation is based on first-order (i.e. six-parameter) flow fields. Shape tracking is performed using two-dimensional radial map representations. The system runs in real-time, and is accurate and reliable. It requires no camera calibration and no knowledge of the camera's motion.     

Automatic Lighting Design using a Perceptual Quality Metric

Lighting has a crucial impact on the appearance of 3D objects and on the ability of an image to communicate information about a 3D scene to a human observer. This paper presents a new automatic lighting design approach for comprehensible rendering of 3D objects. Given a geometric model of a 3D object or scene, the material properties of the surfaces in the model, and the desired viewing parameters, our approach automatically determines the values of various lighting parameters by optimizing a perception-based image quality objective function. This objective function is designed to quantify the extent to which an image of a 3D scene succeeds in communicating scene information, such as the 3D shapes of the objects, fine geometric details, and the spatial relationships between the objects.
Our results demonstrate that the proposed approach is an effective lighting design tool, suitable for users without expertise or knowledge in visual perception or in lighting design.     

Shape blending using the star-skeleton representation

Shape blending, the metamorphosis of one shape into another, is a central problem in two-dimensional computer animation. In spite of impressive uses of morphing in film and video productions, the problem is far from solved. In particular, shape blending still requires considerable manual effort. By decomposing two polygons into equivalent star-shaped pieces and a connecting skeleton, the paper presents a blending method which can represent polygon interiors, not just boundaries, and generate high-quality results with minimal user intervention     

Real-time image marbleization

We present a new real-time image marbleization method that converts an image into a marble-like appearance automatically. The approach models the marbleization process as a two-dimensional fluid dynamics problem, whereby color advection of an input image results in a marbleized image. During the fluid dynamics simulation, we add a pixel-level external force field which is tangent to salient features in the image. The forces are computed from the image characteristics without user intervention. A stylized image with marble-like appearance is easily created that maintains the basic shape of objects in the input image. The entire modeling framework is implemented on a graphics processing unit, thus enabling real-time visual feedback. This approach provides a new tool to design figurative marbling textures without mixing of colors, which are almost impossible with previous computer-generated marbling methods.