Animation of a Virtual Skeleton-based Creature via Motion Retargetting

• Main Authors: Daosheng Mu, 穆道聖
• Other Authors: Tainchi Lu
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/07127535213844772911

碩士 === 國立嘉義大學 === 資訊工程學系研究所 === 96 === In this thesis, we present an efficient approach to performing motion retargetting on a new creature’s skeleton that is generated by taking hierarchical evolution. A meta skeleton is created by means of crossing over two source skeletons. We use mutation to reproduce the next-generation skeleton, in order to perform motion transition in a seamless and smooth manner we operate survivor step to analysis motion data. In addition to the skeleton construction of a new creature by using hierarchical evolution, motion analysis is a required process to preserve the original motion attributes of source objects. Spacetime constraints is taken into account in the proposed system in order to prevent from unexpected artifacts. The proposed technique can effectively produce a new virtual creature’s skeleton with plausible motion data, and are applicable to work for 3D computer games and animations.