Abstract We propose a method that generates stylized building models from examples (Figure ). Our method only requires minimal user input to capture the appearance of a Manhattan world (MW) building, and can automatically retarget the captured ‘look and feel’ to new models. The key contribution is a novel representation, namely the ‘style sheet’, that is captured independently from a building's structure. It summarizes characteristic shape and texture patterns on the building. In the retargeting stage, a style sheet is used to decorate new buildings of potentially different structures. Consistent face groups are proposed to capture complex texture patterns from the example model and to preserve the patterns in the retarget models. We will demonstrate how to learn such style sheets from different MW buildings and the results of using them to generate novel models.

We introduce a video-based approach for producing water surface models. Recent advances in this field output high-quality results but require dedicated capturing devices and only work in limited conditions. In contrast, our method achieves a good tradeoff between the visual quality and the production cost: It automatically produces a visually plausible animation using a single viewpoint video as the input. Our approach is based on two discoveries: first, shape from shading (SFS) is adequate to capture the appearance and dynamic behavior of the example water; second, shallow water model can be used to estimate a velocity field that produces complex surface dynamics. We will provide qualitative evaluation of our method and demonstrate its good performance across a wide range of scenes.

We present a probabilistic approach for the automatic production of tree models with convincing 3D appearance and motion. The only input is a video of a moving tree that provides us an initial dynamic tree model, which is used to generate new individual trees of the same type. Our approach combines global and local constraints to construct a dynamic 3D tree model from a 2D skeleton. Our modeling takes into account factors such as the shape of branches, the overall shape of the tree, and physically plausible motion. Furthermore, we provide a generative model that creates multiple trees in 3D, given a single example model. This means that users no longer have to make each tree individually, or specify rules to make new trees. Results with different species are presented and compared to both reference input data and state of the art alternatives.

Abstract Animating an articulated 3D character requires the specification of its interior skeleton structure which defines how the skin surface is deformed during animation. Currently this task is to a large extent accomplished manually, which consumes a large amount of animators' time. This paper presents an automatic rigging method making use of a new geometry entity called the 3D silhouette. The first step is to extract a coarse 3D curve skeleton and some skeletal joints of a character. This curve skeleton is then refined with a perpendicular silhouette. According to the connectivity of the skeletal joints, the hierarchical animation skeleton is finally constructed. By avoiding complicated computation such as voxelization and pruning, this method is simple and efficient, much faster than existing methods. It proves very useful for quick animation production, with applications including games design and prototype graphical systems. Copyright © 2009 John Wiley & Sons, Ltd.

Abstract A sampled object representation (SOR) defines a graphical model using data obtained from a sampling process, which takes a collection of samples at discrete positions in space in order to capture certain geometrical and physical properties of one or more objects of interest. Examples of SORs include images, videos, volume datasets and point datasets. Unlike many commonly used data representations in computer graphics, SORs lack in geometrical, topological and semantic information, which is much needed for controlling deformation and animation. Hence it poses a significant scientific and technical challenge to develop deformation and animation methods that operate upon SORs. Such methods can enable computer graphics and computer animation to benefit enormously from the advances of digital imaging technology. In this state of the art report, we survey a wide range of techniques that have been developed for manipulating, deforming and animating SORs. We consider a collection of elementary operations for manipulating SORs, which can serve as building blocks of deformation and animation techniques. We examine a collection of techniques that are designed to transform the geometry shape of deformable objects in sampled representations and pay particular attention to their deployment in surgical simulation. We review a collection of techniques for animating digital characters in SORs, focusing on recent developments in volume animation.

Abstract Alpha colours were introduced for image compositing, using a pixel coverage model. Algebraically they resemble homogeneous coordinates, widely used in projective geometry calculations. We show why this is the case. This allows us to extend alpha beyond compositing, to all colour calculations regardless of whether pixels are involved and without the need for a coverage model. Our approach includes multi‐channel spectral calculations and removes the need for 7 channel and 6 channel alpha colour operations. It provides a unified explanation of pre‐multiplied and non pre‐multiplied colours, including negative coordinates and infinite points in colour space. It permits filter and illumination operations. It unifies the three existing significant compositing models in a single framework. It achieves this with a physically‐plausible energy basis. Keywords: projective geometry, homogeneous coordinates, image compositing, alpha blending, alpha compositing, colour representation, filtering, spectral colour, projective alpha colour

Abstract We present a novel image interpolation algorithm. The algorithm can be used in arbitrary resolution enhancement, arbitrary rotation and other applications of still images in continuous space. High‐resolution images are interpolated from the pixel‐level data‐dependent triangulation of lower‐resolution images. It is simpler than other methods and is adaptable to a variety of image manipulations. Experimental results show that the new “mesh image” algorithm is as fast as the bilinear interpolation method. We assess the interpolated images' quality visually and also by the MSE measure which shows our method generates results comparable in quality to slower established methods. We also implement our method in graphics card hardware using OpenGL which leads to real‐time high‐quality image reconstruction. These features give it the potential to be used in gaming and image‐processing applications.

We describe a method for doing image compositing using either 2D geometric shapes or raster images as input primitives. The resolution of the final image is virtually unlimited but, as no frame buffer is used, performance is much less dependant on resolution than with standard painting programs, allowing rendering very large images in reasonable time. Many standard features found in compositing programs have been implemented, like hierarchical data structures for input primitives, lighting control for each layer and filter operations (for antialiasing or defocus).

Putting emotion into figure animation is a difficult task. The paper describes a method towards solving this problem. An emotional model is proposed based on psychological theory and this is integrated into the posturing of the figure. The system is based on general posturing functions which are interpreted depending on the emotional state of the figure. These functions can affect individual joints but are typically used to modify the movement of areas of the body and general stance.

Two-dimensional animation has generally been the poor relation of three-dimensional animation, as far as computer-assistance is concerned. The Animachine project is an attempt to view the whole production process of 2D (cel) animation as ripe for computer-based help, while still allowing artists to retain as much of their traditional way of working as possible. In this paper, we summarize our progress on a renderer for such multi-layer animated graphics. There is little emphasis on the movement aspects of our work. We discuss the various results which can be achieved without recourse to full 3D techniques, we show that the approach is versatile and can be used to achieve a wide range of useful effects.>

We describe a method for modelling cartoon characters, so that the appearance of the character is separate from its movement. We introduce a conceptually simple primitive for modelling the appearance of a character. This primitive is based on NURBS. It is thus very general in what it can model. We also show that it is readily interpolated, for inbetweening. We discuss the interpolation method and the data structures used for appearance and movement. The work is then taken further, to show how movements designed by a traditional animator can be captured. Given a library of appearances, the animator can experiment with finding appropriate movements to match the physical attributes of the character. Moreover, these movements can themselves be blended to produce new movements, giving the animator enhanced creative opportunities. This project is part of our on-going Animachine project, which aims to provide the traditional cel animator with a comprehensive computer-based system.>

2D animation can only be automated to the extent that the computer acts as an interactive assistant to the animator. The key problem is that the 3D information which is implicit in the animated drawings is unavailable and this has encouraged the view of 2D animation being a subset of 3D animation in its fullest generality. However, introducing more than the absolute minimum of 3D information, essentially that contained in a hierarchy of drawing overlays, has matching advantages and disadvantages because animators, for aesthetic reasons, deliberately break the rules of geometry and physics as they apply to real-world objects. A software environment which only supports 2D functionality and drawing overlays is sufficient to promote cost-effectively the full range of effects that animators use while the state of the 3D animation art is as yet incapable of this. Essentially the same techniques can be used on live-capture images once they have had a structure similar to that required for animation imposed on them.

Abstract We describe an accurate method of rendering by scan‐conversion of closed regions bounded by NURBs, with particular comments on its application to computer‐based 2D animation. It is shown that the method is fast, analytically accurate, and can be readily extended to include anti‐aliasing and clipping.

The aim of the Facial Animation, Construction and Editing System (FACES) is to provide a software simulation of the human face. Attention has focused on the face as an important means of non-verbal communication. The interactive composition and modification of the human head and its subsequent animation, have been identified as being of particular interest. The novelty of FACES is that it integrates the modelling and animation of faces using a three-layer anatomical model. FACES consists of four sub-systems: CONSTRUCT, MODIFY, ANIMATE and RENDER. The CONSTRUCT and MODIFY sub-systems enable changes to be made to the structure of the head, at both global and local levels, enabling specific faces to be created. The ANIMATE sub-system caters for motion specification and control, so that real and exaggerated facial expressions can be animated. The RENDER part of the system facilitates the generation of realistic images and their real-time playback. In this paper we consider the system from the user's point of view, examining the facilities which are provided, their appropriateness and practicality.

Many quad encoding technique have been published previously, using a variety of approaches to the different facets of quad storage, representation and manipulation. We review the major classes and identify three parameters which distinguish them. These are then used explicitly to classify a large number of published methods, with short discussion on each. In doing this we are also able to highlight some new approaches, to identify rare examples which escape the classification, and to suggest approaches to choosing a method to suit a given application.

We propose an intuitively simple way of representing colour which has the additional virtue that, it permits mixing and overlaying of transparent and opaque paints to an arbitrary degree. Our approach is related to the earlier alpha channel model used for compositing. It includes this as a special case but has applications in many other areas, especially animation, paint programs and graphics libraries.

We describe an implementation of a networked picture browser. The system offers a pictorial interface to pictorial data, relieves the users of thinking about the underlying filing system, provides managerial tools for installing, moving and deleting pictures, offers graded access and picture sharing and supports a number of picture formats, including hierarchical encodings.

Methods of labelling quadtree nodes by extended quaternary numbers are discussed. It is shown that a fixed bit-length technique requires only 2n+1 bits to represent any node in a depth n quadtree. A technique using exactly that number of bits is described. A more useful extension, requiring 2n+2 bits, is compared with the best technique previously published, which typically requires 2n+4 bits.

Abstract One of the ways in which computer animation can be generated is to use interactive graphics systems. In this paper we describe Controller, an animation system of this type. We concentrate on the operation of Controller and detail how an animator can plan and choreograph the motion of objects and cameras. We have intended to provide a system that gives the animator as much control as possible subject to keeping die interface simple to operate.

Abstract We propose a form of three‐input BitBlt/RasterOp which is useful for pictures constructed in a colour‐mapped form as well as in true colour. The model used for this is that of a piece of potentially multicoloured paper on which the picture appears as ink marks. Computationally simple range testing operations can be applied to the source and destination regions to produce many useful effects, some of which we demonstrate. The resulting operations are primitives which can be used to construct more complex operations.

Abstract Graphical interfaces and interactive graphical programmes are awkward to write because of a lack of top‐down structure. A methodology for constructing graphical programs will be described, together with a system that generates the basic interaction requirements for such applications.

Abstract The ideas underlying a Painting System being developed at the University of Bath are introduced. This differs from a conventional system in that ultra high definition pictures of upto 1 gigabyte may be created. We discuss approaches to implementing a number of standard paint program utilities in a form suited to our high definition method. Finally, we describe an algorithm for the rapid extrusion of brush strokes on a raster device.

Abstract We consider the problem of generating freehand raster graphics pictures at ultra high definition (typically one gigapixel per picture) by scan converting extruded lines. We describe a fast scan conversion algorithm which is linear in final resolution and which incorporates a number of features to ensure high efficiency. Sample pictures and corresponding performance figures are included.

Abstract The authors describe a method of presenting colour pictures which allows users to browse by panning and zooming. The pictures are seen as though on a 512 by 512 by 12 bit framestore, but are defined to 4096 by 4096. The implementation allows the entire picture to be seen in overview, with fine details averaged, or to be zoomed in upon with finer detail progressively revealed by a sequence of twofold linear magnifications. Further zooming, beyond the resolution of the picture data, automatically produces a conventional pixel replication effect. In addition, the picture may be panned vertically or horizontally.

There is considerable scope for the application of computer techniques to help the graphic designer working with printed products, provided that pictures of high technical quality result. Described here is the product of the first stage of a research programme aimed at providing computer assisted drawing aids for high quality colour pictures. The product is an interactive paint program. The paper first explains the nature of the application and the hardware available to the author. It then claims that transcription is central to the task of high quality painting. Next there is a description of the facilities available to the interactive user. Finally, a case study of the implementation of one particular command is used to illustrate the general design of the program and to illustrate why this results in a less rigid interaction than in conventional paint programs.

Summary If we accept the above idiosyncratic views, what kind of architecture will the fifth generation machine offer to graphics users? We would hope to see high quality colour raster graphics, tightly coupled to a sophisticated processor but in its own address space. It should allow for pictures much larger than the screen and also at higher resolution. The local disk must be organized to handle the large amounts of data which picture manipulation on this scale calls for. Disk and main memory must have support for manipulation of two dimensional arrays of pixel data and the filing system and its utilities must reflect this. Response is more important than ever. The software to be built on to this hardware is still a matter for conjecture and depends on whether a particular combination of hardware and language family can be packaged to give clear performance advantages. However a real‐time style operating system is essential. Timescales for suitable architectures seem quite short, around two or three years, with software developing beyond that. Certainly such a machine will contain far more graphical software in the operating system than any before and ought to satisfy a wide range of raster graphics applications.

A multiprocessor architecture for the solution of spatial and graphical problems by recursive subdivision is proposed. The trees that result from such divisions are mapped on to a closed graph, which is then considered as a network of processors. Simulations of these configurations have been performed with encouraging results. Proposals are made for the implementation of such machines.

Most hidden surface algorithms require a considerable amount of computation for all but the simplest images. This prevents their use in real time systems where new frames may be calculated at a rate of 25 per second. The paper presents an approach suitable for fixed models, such as those used in flight simulators, where most of the time consuming computation may be performed when the model is first created. The technique is to test for surface proximity in a well defined manner, ‘well spaced’ surfaces requiring a simple measure of distance to determine priority, and ‘closely spaced’ surfaces being modified until they are ‘well spaced’. This modification is only in the representation of the surface and does not affect its final appearance in the picture. The work to be described is part of a project financed by the Science Research Council, to which grateful acknowledgement is made.