Ken Anjyo

Practice and Theory of Blendshape Facial Models

Blendshapes”, a simple linear model of facial expression, is the prevalent approach to realistic facial animation. It has driven animated characters in Hollywood films, and is a standard feature of commercial animation packages. The blendshape approach originated in industry, and became a subject of academic research relatively recently. This survey describes the published state of the art in this area, covering both literature from the graphics research community, and developments published in industry forums. We show that, despite the simplicity of the blendshape approach, there remain open problems associated with this fundamental technique.

Tour into the picture

“Tour into the Picture” presented by Horry et al. in 1997 is an approach using a single image to create a 3D scene and an animation. Assuming that the picture has one vanishing point, the user specifies a spidery mesh of the scene in the image by establishing that point. That way, perspective projection can be fitted and at most five planes are established. After that, the user models the background, a foreground mask specifying all foreground objects, and also positions the camera. And then it is possible to move the virtual camera by changing its parameters and finally create a short walk-through animation. The technique presented by Kang et al. is based on the original TIP theory. They use vanishing line instead of vanishing point if there are more than one vanishing point or it is hard to find out the vanishing point. Meanwhile they extended the same concept into panoramic images. Institute of omputer raphics, TU Braunschweig C G 1 Seminar Computer Graphics WS 2004/2005 Content Abstract.........................................................................................................................1 Content..........................................................................................................................2

Scattered data interpolation and approximation for computer graphics

The goal of scattered data interpolation techniques is to construct a (typically smooth) function from a set of unorganized samples. These techniques have a wide range of applications in computer graphics. For instance they can be used to model a surface from a set of sparse samples, to reconstruct a BRDF from a set of measurements, to interpolate motion capture data, or to compute the physical properties of a fluid. This course will survey and compare scattered interpolation algorithms and describe their applications in computer graphics. Although the course is focused on applying these techniques, we will introduce some of the underlying mathematical theory and briefly mention numerical considerations.

Scattered data interpolation for computer graphics

The goal of scattered data interpolation techniques is to construct a (typically smooth) function from a set of unorganized samples. These techniques have a wide range of applications in computer graphics and computer vision. For instance they can be used to model a surface from a set of sparse samples, to reconstruct a BRDF from a set of measurements, or to interpolate motion capture data. This course will survey and compare scattered interpolation algorithms and describe their applications in computer graphics. Although the course is focused on applying these techniques, we will introduce some of the underlying mathematical theory and briefly mention numerical considerations.

Sketch interface based expressive hairstyle modelling and rendering

We present a new system for interactively modeling and rendering nonphotorealistic hairstyles. The system is featured with a user-friendly sketch interface allowing a user generate his/her desired hairstyle simply by drawing a few free-form strokes. Hairstyles are modeled with a new polygon based technique called cluster polygon, and can be rendered expressively to obtain hair images similar to those found in cell animation and cartoons

Fluid volume modeling from sparse multi-view images by appearance transfer

We propose a method of three-dimensional (3D) modeling of volumetric fluid phenomena from sparse multi-view images (e.g., only a single-view input or a pair of front- and side-view inputs). The volume determined from such sparse inputs using previous methods appears blurry and unnatural with novel views; however, our method preserves the appearance of novel viewing angles by transferring the appearance information from input images to novel viewing angles. For appearance information, we use histograms of image intensities and steerable coefficients. We formulate the volume modeling as an energy minimization problem with statistical hard constraints, which is solved using an expectation maximization (EM)-like iterative algorithm. Our algorithm begins with a rough estimate of the initial volume modeled from the input images, followed by an iterative process whereby we first render the images of the current volume with novel viewing angles. Then, we modify the rendered images by transferring the appearance information from the input images, and we thereafter model the improved volume based on the modified images. We iterate these operations until the volume converges. We demonstrate our method successfully provides natural-looking volume sequences of fluids (i.e., fire, smoke, explosions, and a water splash) from sparse multi-view videos. To create production-ready fluid animations, we further propose a method of rendering and editing fluids using a commercially available fluid simulator.

Mathematical analysis on affine maps for 2D shape interpolation

This paper gives a simple mathematical framework for 2D shape interpolation methods that preserve rigidity. An interpolation technique in this framework works for given the source and target 2D shapes, which are compatibly triangulated. Focusing on the local affine maps between the corresponding triangles, we describe a global transformation as a piecewise affine map. Several existing rigid shape interpolation techniques are discussed and mathematically analyzed through this framework. This gives us not only a useful comprehensive understanding of existing approaches, but also new algorithms and a few improvements of previous approaches.

A Practical Approach to Direct Manipulation Blendshapes

Abstract The original approach of direct manipulation blendshapes introduced a mathematical framework for manipulating blendshape face models directly and interactively while interoperating with traditional slider-based editing. This article introduces a few practical and useful techniques to the original work and presents the implementation details as a software tool. In addition, this article describes a statistical-prior algorithm that “learns” from previous animation and allows rapid creation of expressive animation in a given style. The animation examples illustrate how effectively this tool works for practical use.

Cartoon hair animation based on physical simulation

The hair movement in cel character animation is sometimes inconsistent. The representation of hair, therefore, is a specialized work in cel animation. It is difficult to achieve in computer graphics, since cel animation may not be consistent with all the hair’s attributes from all camera positions. In fact, what animators create by solving physics equations is not always what the animators desire even though it is physically correct. Also the motion of hair in cartoon animation carries meaning. Therefore, what the animators want exists only in their imagination, and all the hair forms (shape modeling and number of hairs) in the key frames from the animation sequence cannot be in complete agreement (Fig1). Quite convincing results, however, can be obtained from physically inconsistent frames. This is the most difficult part of expressing cartoon hair animation in computer graphics and is the reason why cartoon hair animation has been done by hand. It requires the instincts of an expert animator – and is very time consuming. In fact, the work of even a skilled animator is highly demanding. Yet there have been little researches[1] directed to the solutions to time-consuming problems such as the cel animation of hair motion.

Mathematical Formulation of Motion and Deformation and Its Applications

This chapter is intended to give a summary of Lie groups and Lie algebras for computer graphics, including an example from interpolations and blending of motions and deformation. In animation and filmmaking procedure, we want to get a smooth transition of the drawing of given starting and ending point (the drawings at these points are called key frame). This problem can be understood as an interpolation, so that various approach have been proposed and used in Computer Graphics. After the seminal work so-called ARAP (as-rigid-as-possible), serious properties of matrix groups have been taken into account both in theoretical and in computational point of view. To understand and develop these properties, we here employ a Lie theoretic approach.