New ETHZ masters thesis project available: Inverse kinematics and procedural animation for stretchable and twistable bones
Alec Jacobson
January 01, 2012
Olga Sorkine and I will be hosting a master's thesis project. The project, entitled Inverse kinematics and procedural animation for stretchable and twistable bones is now available, and we are eagerly awaiting applications. For many years, animators have used internal skeletons to drive 3D character deformation. These skeletons are usually composed of rigid bones, which do not change length. This requirement allows for efficient algorithms for creating real-time animations. Further, it facilitates inverse kinematics and procedural animation. Inverse kinematics is the problem of optimizing the positions and orientations of each bone to place the end effectors (e.g. hands and feet) in a user-specified configuration. Procedural animation automatically animates bones of a skeleton to achieve realistic animations with minimal user directing.
Often in cartoon animation our characters should stretch, perhaps in a very exaggerated way. If we use a skeleton then stretching means bones must change length, breaking compatibility with existing inverse kinematics and procedural animation techniques. Likewise, tradition skeleton-based animation limits twisting to joints, where bones meet. Twisting the character along the length of a single bone is not supported by these traditional methods.
In this project, we will explore new inverse kinematics and procedural animation methods for skeleton-based deformation where bones may stretch and twist. These new methods will generalize existing ones. However, part of our challenge will be to design them efficiently so as to highlight their power over past techniques without losing speed. The first step is to quickly learn how traditional skeleton-based skinning works, via implementation. The next step is to learn how traditional inverse kinematics and procedural animations works, via implementation, meanwhile building an intuition as to how stretchable, twistable bones could fit in or not. Finally, we will design and implement a generalized inverse kinematics or procedural animation technique which allows bones to stretch and twist. Please don't hesitate to contact me for more details.
Also, check out the full list of IGL projects.
Note: You will need to be at an ETH IP address to visit these links.