Style translation for human motion
Eugene Hsu, Kari Pulli, Jovan Popović
In ACM Transactions on Graphics, 24(3), August 2005.
Abstract: Style translation is the process of transforming an input motion into a new style while preserving its original content. This problem is motivated by the needs of interactive applications, which require rapid processing of captured performances. Our solution learns to translate by analyzing differences between performances of the same content in input and output styles. It relies on a novel correspondence algorithm to align motions, and a linear time-invariant model to represent stylistic differences. Once the model is estimated with system identification, our system is capable of translating streaming input with simple linear operations at each frame.
Keyword(s): data mining, human simulation, machine learning
Article URL: http://doi.acm.org/10.1145/1073204.1073315
BibTeX format:
@article{Hsu:2005:STF,
  author = {Eugene Hsu and Kari Pulli and Jovan Popović},
  title = {Style translation for human motion},
  journal = {ACM Transactions on Graphics},
  volume = {24},
  number = {3},
  pages = {1082--1089},
  month = aug,
  year = {2005},
}
Search for more articles by Eugene Hsu.
Search for more articles by Kari Pulli.
Search for more articles by Jovan Popović.

Return to the search page.


graphbib: Powered by "bibsql" and "SQLite3."