We present a method for creating black-and-white illustrations from photographs of human faces. In addition an interactive technique is demonstrated for deforming these black-and-white facial illustrations to create caricatures which highlight and exaggerate representative facial features. We evaluate the effectiveness of the resulting images through psychophysical studies to assess accuracy and speed in both recognition and learning tasks. These studies show that the facial illustrations and caricatures generated using our techniques are as effective as photographs in recognition tasks. For the learning task we find that illustrations are learned two times faster than photographs and caricatures are learned one and a half times faster than photographs. Because our techniques produce images that are effective at communicating complex information, they are useful in a number of potential applications, ranging from entertainment and education to low bandwidth telecommunications and psychology research. Categories and Subject Descriptors: I.3.3 [Computer Graphics]: Picture/image Generation—bitmap and framebuffer operations;

Constraints have been traditionally used for computer animation applications to define side conditions for generating synthesized motion according to a standard, usually physically realistic, set of motion equations. The case of facial animation is very di#erent, as no set of motion equations for facial expressions is available. In this paper we discuss a facial animation editor, which uses numerical constraints for two roles: to declare the mimic repertoire of synthetic faces and other requirements a facial animation has to meet, and to aid the animator in the process of composing a specific animation fulfilling the requirements. The editor is thus also a `motion sculpturing` tool, which lifts the task of creating facial animation from the control data manipulation level to the conceptual design level. The major aid of the editor is to repair inconsistencies due to changes made by the user, and revise changes for which no good repair is possible. Also, reuse of constrained animations, especially expressions, is supported. The main machinery behind these services is interval propagation, which, if using certain type of linear inequalities to express the character- as well as the animation-specific requirements, can produce quickly the interval of feasible values for each control variable. If a solution (usually, repair) has to be produced, it is generated by selecting the best one from a restricted set of acceptable solutions, based on user-defined or automatically generated criteria for the choices.

We discuss the CharToon system by which one can interactively construct parametrized 2D drawings and a set of time curves to animate the drawings. The system has been implemented in JAVA and used to produce and animate 2D cartoon faces. The variety and potential usage of such faces is to illustrated by a series of demos. Keywords: animation, facial modeling, JAVA, constraints. 1. Introduction Computer facial animation has been a flourishing research topic for more than 25 years, aiming at building models which can be animated and used to (re-)produce facial expressions reflecting emotions and mouth movements for spoken text 7 . The bulk of the efforts has been spent on producing 3D facial models, which both in appearance and in facial movements match faithfully the real human face. In spite of the extensive research, it is still an unsolved problem to achieve full realism on synthetic faces. Besides difficulties with modeling details such as hair and skin texture, it has been emp...

In this paper, we present a new interactive 2D deformation technique to make caricatures with extreme exaggerations. In our technique, we use simplices as deformation primitives. User defined source and target shapes of each simplex primitive define local coordinates. These local coordinates determine a unique translation vector for any given point in 2D space. We also provide a set of blending functions to effectively interpolate these translation vectors from each simplex primitive to compute a combined translation. These combined translation vector are used to transform each vertex of a texture mapped grid. Deformation of the grid provide interactive deformation of the texture that is mapped to the grid.

Facial expressions are becoming more and more important in today’s computer systems with humanoid user interfaces. Avatars have become popular, however their facial communication is usually limited. This is partly due to the fact that many questions, especially on the dynamics of expressions, are still open. Moreover, the few commercial facial animation tools have limited facilities, and are not aimed at lightweight Web applications. In this article we discuss the empirical basis and a software tool to produce faces with emotional expressions and lip-sync. In order to elicit the characteristics of expressions on real human faces and to map them on synthetic, non-realistic ones, we analysed expressions on real and artist-drawn cartoon faces. We developed CharToon, a software tool that allows the construction of faces and to animate them from scratch or by re-using components of the facial feature and expression repertoire. CharToon faces can be animated real time. The list of applications includes 3D faces of avatars in a VRML environment.

As a computational bridge between the high-level a priori knowledge of object shape and the low-level image data, active contours (or snakes) are useful models for the extraction of deformable objects. We propose an approach for manipulating multiple snakes iteratively, called interacting snakes, that minimizes the attraction energy functionals on both contours and enclosed regions of individual snakes and the repulsion energy functionals among multiple snakes that interact with each other. We implement the interacting snakes through explicit curve (parametric active contours) representation in the domain of face recognition. We represent human faces semantically via facial components such as eyes, mouth, face outline, and the hair outline. Each facial component is encoded by a closed (or open) snake that is drawn from a 3D generic face model. A collection of semantic facial components form a hyper-graph, called semantic face graph, which employs interacting snakes to align the general facial topology onto the sensed face images. Experimental results show that a successful interaction among multiple snakes associated with facial components makes the semantic face graph a useful model for face representation, including cartoon faces and caricatures, and recognition.

The aim of this study was to evaluate category learning in schizophrenia on tests of perceptual abstraction. Participants learned to categorize simple geometrical shapes. The categories were either well-defined (discrete categories, or DCs) or ill-defined (graded categories,

committee and by majority vote has been found to be satisfactory.

A local smoothing operator applied in the reverse direction is used to obtain planar shape enhancement and exaggeration. Inversion of a smoothing operator is an inherently unstable operation. Therefore, a stable numerical scheme simulating the inverse smoothing effect is introduced. Enhancement is obtained for short time spans of evolution. Carrying the evolution further yields shape exaggeration or caricaturization effect. Introducing attraction forces between the evolving shape and the initial one, yields an enhancement process that converges to a steady state. These forces depend on the distance of the evolving curve from the original one and on local properties. Results of applying the unrestrained and restrained evolution on planar shapes, based on a stabilized inverse Geometric Heat Equation, are presented showing enhancement and caricaturization effects. 1 Introduction In this paper we consider possible ways to design an automatic procedure for enhancing and caricaturizing plan...

In this paper, we present a method for automatically creating black-and-white drawings of human faces from photographs. We then demonstrate new techniques for deforming these drawings to create caricatures intended to highlight and exaggerate facial features. A number of psychophysical studies were performed that show that both the black-and-white drawings and the caricatures generated using our techniques have similar effectiveness as photographs in recognition tasks, and perform better than photos in learning tasks. Our methods are useful for a number of potential applications, ranging from forensics to telecommunications and from cartoons to education and psychology research.