We present a technique to extract motion parameters of a human figure from a single video stream. Our goal is to prototype motion synthesis rapidly for game design and animation applications. For example, our approach is especially useful in situations where motion capture systems are restricted in their usefulness given the various required instrumentation. Similarly, our approach can be used to synthesize motion from archival footage. By extracting the silhouette of the foreground figure and using a model-based approach, the problem is re-formulated as a local, optimized search of the pose space. The pose space consists of 6 rigid body transformation parameters plus the internal joint angles of the figure. The silhouette of the figure from the captured video is compared against the silhouette of a synthetic figure using a pixel-by-pixel, distance-based cost function to evaluate goodness-of-fit. For a single video stream, this is not without problems. Occlusion and ambiguities arising from the use of a single view often cause spurious reconstruction of the captured motion. By using temporal coherence, physical constraints, and knowledge of the anatomy, a viable pose sequence can be reconstructed for many live-action sequences.

We present a facial model designed primarily to support animated speech. Our facial model takes facial geometry as input and transforms it into a parametric deformable model. The facial model uses a muscle-based parameterization, allowing for easier integration between speech synchrony and facial expressions. Our facial model has a highly deformable lip model that is grafted onto the input facial geometry to provide the necessary geometric complexity needed for creating lip shapes and high-quality renderings. Our facial model also includes a highly deformable tongue model that can represent the shapes the tongue undergoes during speech. We add teeth, gums, and upper palate geometry to complete the inner mouth. To decrease the processing time, we hierarchically deform the facial surface. We also present a method to animate the facial model over time to create animated speech using a model of coarticulation that blends visemes together using dominance functions. We treat visemes as a dynamic shaping of the vocal tract by describing visemes as curves instead of keyframes. We show the utility of the techniques described in this paper by implementing them in a text-to-audiovisual-speech system that creates animation of speech from unrestricted text. The facial and coarticulation models must first be interactively initialized. The system then automatically creates accurate real-time animated speech from the input text. It is capable of cheaply producing tremendous amounts of animated speech with very low resource requirements.

Abstract We describe techniques used to create animations of song. Modifications to a text‐to‐audiovisual‐speech system have been made to take the extra information of timing and frequency of the lyrics from a MIDI file. Lip‐synchronized animations of song are then produced. We discuss differences between the production of speech and the production of song. Copyright © 2004 John Wiley & Sons, Ltd.

We describe techniques used to create animations of song. Modifications to a text-to-audiovisual-speech system have been made to take the extra information of timing and frequency of the lyrics from a MIDI file. Lip-synchronized animations of song are then produced. We discuss differences between the production of speech and the production of song.

Interactive aesthetic selection is used to evolve populations of images. The images are generated using continuous layered pattern functions which are based on procedural texturing techniques. Gene values are used to control the visual properties of regular and irregular layered feature patterns. A wide range of feature shapes is produced by implicit shape function blending and bombing. The evolved images are used to produce height fields which can be converted to multiuser game maps. The parametric design space defined by the chosen representation can be controllably biased toward specific formal qualities.

Lip-sync animation is complex and challenging. It promises to be important in natural human-computer interfaces and entertainment as well as aid in the education of the deaf. It is an important component in creating a realistic human figure. Speech is based on principles from anatomy, physics, and psychophysiology. We discuss some of the issues that make speech so complex to model visually.

Abstract We present a tongue model for use in computer‐animated speech. The model is capable of representing tongue shapes during the production of English vowels and consonants as well as general motion of the tongue. Geometrically, the model is composed of a B‐spline surface with 60 control points and an 8×13 grid of bi‐cubic patches. We also present a parameterization of the model that requires only six parameters for use during speech production. This parameterization can also be used to define the deformation of other tongue models. Copyright © 2001 John Wiley & Sons, Ltd.

Computer animation has a history as long as computer graphics. It's old enough to have interesting divisions, blind alleys, and new trends. It's young enough for the new entrants to have the opportunity to meet most of the people who started the field. This paper reviews the historical development of computer animation noting the trends and suggesting the future.

This paper describes an algorithm for automatically generating a control skeleton for use in animating a polygonal data model. The algorithm has several steps. First, the figure is voxelized, and a Euclidean distance map is computed for voxels interior to the figure. A path creation algorithm then generates a tree-structured set of voxel paths spanning the object's interior. Simplification of these paths leads to a control skeleton for the object. The voxelization is used again to anchor the vertices of the polygonal model to the skeleton. Afterwards, new joint angles may be specified for the skeleton, and once the skeleton has been updated, vertices of the model can be repositioned accordingly. Unlike previous methods, the algorithm is fully automated, requiring very little user input. Also, it can be applied successfully to a wide variety of objects. Furthermore, the algorithm is fast-it can produce a useful control skeleton for most objects in under two minutes on a low-end PC.

Anatomy increases the sensitivity of the artist's eye and makes the skin transparent; it allows the artist to grasp the true form of the surface contours of the body because he knows the parts that lie hidden beneath a veil of flesh."Gerdy

Abstract Task level animation of articulated figures, such as the human body, requires the ability to generate collision‐free goal‐directed motion of individual limbs in the presence of obstacles. This paper describes a new articulated limb motion planner for goal‐directed point‐to‐point reaching motions. The produced motion avoids obstacles while optimizing an objective function. This two‐phase algorithm uses heuristic guided Monte Carlo techniques to create a consistent underlying paradigm. The first phase consists of an existing potential field based random path planner which generates a population of candidate paths. This initial population is fed into the second phase, a genetic algorithm, which iteratively refines the population as it optimizes with respect to the objective function. The refinement process works on the principle ofpath coherency, the idea that a family of closely related collision‐free paths lies in the vicinity of a given collision‐free path. This paper focuses on seven different optimization functions. Optimized trajectories produced by the new motion planner are compared to those generated solely by the random path planner. The presented algorithm is flexible in that a wide range of objective functions can be optimized. Applications of the algorithm include task level animation, ergonomics and robotics.

Techniques that transform one two-dimensionaf image into another have gained widespread use m recent yeara.Extending these techniques to transform pairs of 3D objects, as opposed to 2D images of the objects, providea several advsntagea, including the ability to snimate the objects independently of the transformation.This paper presents an algorithm for computing such transformations.The algorithm merges the topological structures of a pair of 3D polyhedral models into a common vertex/edgeJface network.This allows trsmsformations from one object to the other to be easily computed by interpolating between corresponding v@ex positions.

Abstract The issues relating to the shape transformation problem are discussed and a new algorithm is presented for computing the transformation of one shape into another. In this algorithm, the boundary definitions of the two initial shapes are used and a mapping is established between the vertices and edges of the respective objects. New vertices and edges are introduced into the object definitions when necessary to establish a one‐to‐one vertex correspondence and to match connectivity relationships between vertices. These can then be used to do a vertex‐to‐vertex interpolation that maintains valid polyhedral topologies for all of the intermediate shapes. The algorithm establishes a mapping between areas of the object such that adjacency relationships are preserved. These areas are recursively subdivided so that adjacency relationships of subareas are also preserved. During subdivision, vertices and edges are added to the boundaries of subareas so that a one‐to‐one mapping is established between them. Subdivision continues until each subarea consists of a single face. The algorithm presented works for objects that are topologically equivalent to spheres and can easily be extended to other pairs of objects as long as they are topologically equivalent to each other.

This paper describes a new technique that efficiently combines volume rendering and scanline a-buffer techniques. This technique is useful for combining all types of volume-rendered objects with scanline rendered objects and is especially useful for rendering scenes containing gaseous phenomena such as clouds, fog, and smoke. The rendering and animation of these phenomena has been a difficult problem in computer graphics.A new algorithm for realistically modeling and animating gaseous phenomena is presented, providing true three-dimensional volumes of gas. The gases are modeled using turbulent flow based solid texturing to define their geometry and are animated based on turbulent flow simulations. A low albedo illumination model is used that takes into consideration self-shadowing of the volumes.

A methodology is proposed for creating and animating computer generated characters which combines recent research advances in robotics, physically based modeling and geometric modeling. The control points of geometric modeling deformations are constrained by an underlying articulated robotics skeleton. These deformations are tailored by the animator and act as a muscle layer to provide automatic squash and stretch behavior of the surface geometry. A hierarchy of composite deformations provides the animator with a multi-layered approach to defining both local and global transition of the character's shape. The muscle deformations determine the resulting geometric surface of the character. This approach provides independent representation of articulation from surface geometry, supports higher level motion control based on various computational models, as well as a consistent, uniform character representation which can be tuned and tweaked by the animator to meet very precise expressive qualities. A prototype system (Critter) currently under development demonstrates research results towards layered construction of deformable animated characters.

A computer animation system is discussed which employs interactive techniques and presents a unified approach to the graphical display of complex three dimensional data. The system facilitates the generation, manipulation and display of highly detailed data with the aid of interactive devices and a video interface to a standard color TV monitor. The system enables the animator to create a variety of objects (including texture) and to specify the necessary transformations for animation sequences. A run length processing technique combined with a brute force Z-buffer algorithm has been newly designed and implemented that can handle the intersection of several million faces, lines and points. This makes possible a full range of visual cues to simulate fire, smoke, water and complex 3-D texture such as grass, hair and bark. Basic concepts and approaches are described. The display algorithm and the procedure model to generate texture are presented and the implications of the system for computer animation are discussed. Extensions to the system are outlined which include a unique graphics display processor currently under construction that includes a partial implementation of the display algorithm in hardware.

A major research area in 3-D computer graphics is the inputting of complex descriptions. This paper describes our current attempt at solving that problem: creation of a sculptor's studio-line environment in which the user is provided with various tools to shape, cut and join objects. The emphasis of the implementation has been on naturalness and habitability. The issues involved in designing such a system, especially in a minicomputer-based color raster-scan animation environment, are discussed. The basic algorithms are described in some detail and a fast efficient implementation of a hidden-line algorithm is explained.