Zhihua Chen, Shengxi Wu, Hong Zheng, Weiwen Meng, Xin Xu
{"title":"Creating Realistic Human Motions with PBSO","authors":"Zhihua Chen, Shengxi Wu, Hong Zheng, Weiwen Meng, Xin Xu","doi":"10.1109/VIZ.2009.51","DOIUrl":null,"url":null,"abstract":"It is a difficult task to create a realistic human animation because of the high complexity of human motion. To address this problem, a new method is presented for producing physically valid motion with example motions. The core of our method is physics-based space-time optimization (PBSO). PBSO introduces physical constraints into conventional space-time optimization and then ensure the physical reality of the edited motion. The intermediate model is employed to solve the convergence of optimization process. The original motion is transformed into the simplified motion and then edited. Experiments shows the captured motion data can be correctly copied to a new character or new path with the approach.","PeriodicalId":315752,"journal":{"name":"2009 Second International Conference in Visualisation","volume":"52 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2009-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2009 Second International Conference in Visualisation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/VIZ.2009.51","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
It is a difficult task to create a realistic human animation because of the high complexity of human motion. To address this problem, a new method is presented for producing physically valid motion with example motions. The core of our method is physics-based space-time optimization (PBSO). PBSO introduces physical constraints into conventional space-time optimization and then ensure the physical reality of the edited motion. The intermediate model is employed to solve the convergence of optimization process. The original motion is transformed into the simplified motion and then edited. Experiments shows the captured motion data can be correctly copied to a new character or new path with the approach.