DocumentCode
2688550
Title
An Avatar Motion Generation Method Based on Inverse Kinematics and Interactive Evolutionary Computation
Author
Akase, Ryuya ; Nishino, Hiroaki ; Kagawa, Tsuneo ; Utsumiya, Kouichi ; Okada, Yoshihiro
Author_Institution
Dept. of Comput. Sci. & Intell. Syst., Oita Univ., Oita, Japan
fYear
2012
fDate
4-6 July 2012
Firstpage
741
Lastpage
746
Abstract
This paper presents an automatic animation generation method for 3DCG humanoid avatars. We can create 3DCG animation even at the level of individuals by using commercially available software tools. Creating 3DCG animation, however, demands a variety of expertise and a mastery of 3DCG authoring tools. Therefore, people may easily be bothered for creating 3DCG animation contents. To solve the problem, we designed and developed a system allowing even novices to create 3DCG animation contents without special knowledge and programming skills. The proposed method uses Inverse Kinematics (IK) and Interactive Evolutionary Computation (IEC) for readily designing and generating avatar motions. The system only requires users to indicate some evaluation values and creates their preferable avatar motions based on the specified values. Additionally, we propose a new operation method to reduce the users´ burden in an IEC-based 3DCG contents generation.
Keywords
avatars; computer animation; evolutionary computation; humanoid robots; robot kinematics; software tools; 3DCG authoring tool; 3DCG humanoid avatar; IEC; IK; automatic animation generation method; interactive evolutionary computation; inverse kinematics; software tool; Animation; Avatars; Gene expression; IEC; Joints; Trajectory; 3DCG animation; Interactive Evolutionary Computation; Inverse Kinematics; human computer interaction; humanoid avatar;
fLanguage
English
Publisher
ieee
Conference_Titel
Complex, Intelligent and Software Intensive Systems (CISIS), 2012 Sixth International Conference on
Conference_Location
Palermo
Print_ISBN
978-1-4673-1233-2
Type
conf
DOI
10.1109/CISIS.2012.124
Filename
6245663
Link To Document