New Techniques for Interactive Character Animation
Artificial Intelligence/Machine Learning
Gaming & Interactive
Research & Education
DescriptionThe application of deep learning for physics-based character animation and cinematic controllers for interactive animation is changing how we should think about interactive character animation in video games and virtual reality. We review the benefits and drawbacks of the techniques used and the implementations available to get started.
Prerequisites A detailed understanding of quaternion algebra is not needed to understand the main ideas of this course. However, to understand the technical detail of the implementations behind these solutions, it is necessary to be comfortable with these operations. We also recommend the person interested in such field to get acquainted with the twist-swing decomposition. Several sources for those materials can be found online, a possible way to learn those systematically is going through the course materials and exercises that can be found here: https://joanllobera.github.io/teaching/animation-foundations
Intended Audience This course is designed for software engineers who want to start exploring how machine learning techniques for character animation work. It also can be beneficial for creative minds who want to use these techniques by adapting existing implementations without having to dig too deep in the technical details.