Style Translation for Human Motion
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 differen...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Published: |
2008
|
| Online Access: | http://hdl.handle.net/1721.1/41945 |
| _version_ | 1826195970753298432 |
|---|---|
| author | Hsu, Eugene Pulli, Kari Popovic, Jovan |
| author2 | Jovan Popovic |
| author_facet | Jovan Popovic Hsu, Eugene Pulli, Kari Popovic, Jovan |
| author_sort | Hsu, Eugene |
| collection | MIT |
| description | 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. |
| first_indexed | 2024-09-23T10:18:44Z |
| id | mit-1721.1/41945 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T10:18:44Z |
| publishDate | 2008 |
| record_format | dspace |
| spelling | mit-1721.1/419452019-04-05T16:19:02Z Style Translation for Human Motion Hsu, Eugene Pulli, Kari Popovic, Jovan Jovan Popovic Computer Graphics 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. 2008-08-25T19:00:49Z 2008-08-25T19:00:49Z 2005-08-01 http://hdl.handle.net/1721.1/41945 ACM Transactions on Graphics, 24(3):1082-1089, August 2005. Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory N/A application/octet-stream |
| spellingShingle | Hsu, Eugene Pulli, Kari Popovic, Jovan Style Translation for Human Motion |
| title | Style Translation for Human Motion |
| title_full | Style Translation for Human Motion |
| title_fullStr | Style Translation for Human Motion |
| title_full_unstemmed | Style Translation for Human Motion |
| title_short | Style Translation for Human Motion |
| title_sort | style translation for human motion |
| url | http://hdl.handle.net/1721.1/41945 |
| work_keys_str_mv | AT hsueugene styletranslationforhumanmotion AT pullikari styletranslationforhumanmotion AT popovicjovan styletranslationforhumanmotion |