Kyle Reed will be presenting his work with Prof Darren Cosker at the ACM SIGGRAPH/Eurographics Symposium of Computer Animation (SCA). This prestigious event has been the premier forum for innovations in the software and technology of computer animation since 2002.

We caught up with Kyle to hear more about his paper:

User‐Guided Facial Animation through an Evolutionary Interface

Prototyping is vital for communicating designs for ambiguous tasks, such as creating characters with atypical anatomies or when deviating from an actor’s performance. As this is hard to define numerically, conventional methods rely on animator intuition to manually adjust the many parameters of a face rig. This is not only labour-intensive, but also subjective as might not comply with a creative stakeholder’s vision.

We propose a design framework that streamlines prototyping tasks in facial animation – without requiring any ground truth reference or animation experience. Our approach encodes an Evolutionary Algorithm driven by selection from arbitrary users as a search heuristic. Presented as a simple interface, users sample control combinations and select favourable results to influence later sampling. Over multiple iterations of disregarding unfavourable parameter values, parameters converge towards the user’s ideal.

Aside from animation, such a tool could be particularly useful for designing facial perception studies in psychology and the social sciences. The tool is currently being trialled in a perception study for individuals with psychiatric disorders. Furthermore, as our evolutionary interface treats the model as a black-box function, our design is easily adapted to various parametric models that can be assessed through appearance. Notable applications include exploring latent variable spaces of generative models and configuration settings for tasks with many parameters such as rendering.

You can read the full paper here.

Reed, K., & Cosker, D. (2019). User-Guided Facial Animation through an Evolutionary Interface. Computer Graphics Forum. https://doi.org/10.1111/cgf.13612