Virtual Human Gesture Prediction Using Machine Learning

A big problem in animation lies in the animation of hands and fingers, which is normally done manually and is a known hassle. With this in mind, I attempt to use machine learning to train an algorithm that can correctly classify body motions to the corresponding gesture correctly. After parsing the clips’ keyframes I use rotations and positions of the various body parts to create features that refer to the gesture in total. I attempt a variety of machine learning methods in order to attempt to achieve the best results.

Speaker(s)

, Student, Santa Clara University