View-Dependent Character Animation

ISBN-13: 9781447158837 / Angielski / Miękka / 2014 / 142 str.

Parag Chaudhuri; Prem Kalra; Subhashis Banerjee

Opis

Computer generated 3D animation has matured over the years into a complex art form. Coordinating and presenting the character's movement in three dimensions to convey a specific idea to the audience requires artistic and technical skills, and often a labourious iterative trial-and-error process to get it right. Creating moving camera character animations in 3D is a multi-faceted computer graphics and computer vision problem. Warranting a formal representation of the moving camera, and efficient algorithms to help author the multitude of character poses required for the animation. It is also necessary to deal with issues pertaining to camera and character pose interpolation and visualization of the association between the two. The solution to this problem has to be efficient and elegant from the perspective of a computer scientist, and make sense and be intuitive to use. This well-researched book contains a large number of example animations to explain and illustrate this versatile technique.

Recenzje

From the reviews:

"This book report on a view-dependent approach as a means of accomplishing this task in a reliable manner. ... Three very important appendices are provided. ... This book targets a wide audience, including both the academic research and the professional animation artist. It attempts to be self-contained ... . A number of figures are provided to help the reader understand the concepts presented. ... This book is comprehensive, and is a job well done." (R. Goldberg, Computing Reviews, January, 2008)

Spis treści

Introduction.- Principles of Animation.- The Animation Pipeline.- Moving-camera Character Animation.- Designing a Framework for View-Dependant Animation.- Tour of the Book.- A Framework for View-Dependant Animation.- Prior Work.- The View Space.- Distance of Viewpoint.- Other Extensions.- Chapter Summary.- View-Dependant Animation from Sketches.- Prior Work.- Overview of the Pipeline.- Inputs.- Recovering the Camera.- Posing the Character.- Animating the Character.-Discussion of Other Results.- Chapter Summary.-View-Dependant Animation from Multi-modal Inputs.- Challenges in Multi-Modal Authoring of Animation.- Prior Work.- Creating a View Space from Video.- Creating the View Space from Multi-modal Inputs.- Generating the Animation from Video.- Generating the Animation from Multi-Modal Inputs.- Chapter Summary.- Stylistic Reuse of View-Dependant Animations.- Prior Work.- Animating Multiple Characters from the Same View Space.- Animating Multiple Characters from Multiple View Spaces.- Animating Different Parts of a Single character from a Single View Space.- Chapter Summary.- Discussion and Future Directions.- Discussion.- Future Directions.- Camera Models and Computation of the Camera Matrix.- The Exponential Map Parameterization of Rotations.-Spherical Joint Limits with Reach Cones.-References.- Index

Blurb

Computer generated 3D animation has matured over the years into a rich and complex art form. The need to animate diverse characters has caused character animation to become an extensively researched area. Coordinating and presenting the character's movement in three dimensions to convey a specific idea to the audience, however, still remains a challenge. The animator has to employ a lot of artistic and technical skills, and often a labourious iterative trial-and-error process to get it right. It is even more difficult for the animator to generate the character's action if the point of view (i.e., the rendering camera) is also varying. Creating moving camera character animations in 3D is, however, a multi-faceted computer graphics and computer vision problem. It warrants a formal representation of the moving camera, and efficient algorithms to help author the multitude of character poses required for the animation. It is also necessary to deal with issues pertaining to camera and character pose interpolation and visualization of the association between the two. The solution to this problem has to be efficient and elegant from the perspective of a computer scientist, yet at the same time make sense and be intuitive to use.

This well-researched book develops and demonstrates a framework for view-dependent animation designed to solve these issues, well illustrated with a large number of example animations to explain and illustrate this versatile technique.