Abstract |
This thesis presents a framework for intuitive and efficient creation of moving camera character animations. In particular, it presents a unified theoretical framework for view-dependent character animation, where the characters action depends on the view. The framework can represent and use all variations possible in defining cameras. It can extract camera and character pose information from multi-modal inputs like sketches and videos, in order to create a view space. It then allows a quick exploration of the various animations which can be generated from this space. It can be easily extended to include other forms of input such as motion capture. The dissertation also explores how it is possible to reuse the view-dependent variations to animate multiple characters. |
Animation Videos |
All the animations are encoded using the standard DivX codec (available from http://www.divx.com). The clips can be played using mplayer on Linux or Windows Media Player on Windows or any other player which understands the DivX format. |
Related Publications |
|
Page last updated on 28 July, 2006. |
![]() |
© Parag Chaudhuri , 2009 |
![]() |
![]() |
![]() |
|