Avatar style transformation using neural networks

A system comprising:

• a processor configured to:
• retrieve, from a storage device, an avatar depicting motion in a first style;
• receive user input selecting a second style;
• obtain, based on the user input, a first trained machine learning model that performs a non-linear transformation of motion from the first style to the second style, wherein a plurality of trained machine learning models is stored that each performs a different non-linear transformation of motion from one style to another style, and wherein the plurality of stored machine learning models includes at least the first trained machine learning model that performs the non-linear transformation of motion from the first style to the second style and a second trained machine learning model that performs a non-linear transformation of motion from the first style to a third style;
• and
• apply the obtained first trained machine learning model to the retrieved avatar to transform the avatar from depicting motion in the first style to depicting motion in the second style.

A non-transitory machine-readable storage medium that includes instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:

• retrieving, by one of the processors from a storage device, an avatar depicting motion in a first style;
• receiving user input selecting a second style;
• obtaining, based on the user input, a first trained machine learning model that performs a non-linear transformation of motion from the first style to the second style, wherein a plurality of trained machine learning models is stored that each performs a different non-linear transformation of motion from one style to another style, and wherein the plurality of stored machine learning models includes at least the first trained machine learning model that performs the non-linear transformation of motion from the first style to the second style and a second trained machine learning model that performs a non-linear transformation of motion from the first style to a third style;
• and
• applying the obtained first trained machine learning model to the retrieved avatar to transform the avatar from depicting motion in the first style to depicting motion in the second style.