System and method for generating character poses using deep learning
AssigneeUnity IPR ApS
InventorsFlorent Benjamin Bocquelet, Dominic Laflamme, Boris Oreshkin
Life of this patent
Estimating the term…
Abstract
A method of generating or modifying poses in an animation of a character are disclosed. Variable numbers and types of supplied inputs are combined into a single input. The variable numbers and types of supplied inputs correspond to one or more effector constraints for one or more joints of the character. The single input is transformed into a pose embedding. The pose embedding includes a machine-learned representation of the single input. The pose embedding is expanded into a pose representation output. The pose representation output includes local rotation data and global position data for the one or more joints of the character.
Claims17
A system comprising:
- one or more computer processors;
- one or more computer memories;
- a set of instructions stored in the one or more computer memories, the set of instructions configuring the one or more computer processors to perform operations, the operations comprising:
- combining variable numbers and types of supplied inputs into a single input, the variable numbers and types of supplied inputs corresponding to one or more effector constraints for one or more joints of a character;
- transforming the single input into a pose embedding, the pose embedding being a machine-learned representation of the single input;
- and
- expanding the pose embedding into a pose representation output, the pose representation output including local rotation data and global position data for the one or more joints of the character, wherein the expanding of the pose embedding into a pose representation output includes translating the pose embedding into unconstrained predictions of 3D joint positions for the one or more joints, the translating using a global position decoder machine learned model, and generating the local rotation angles for the one or more joints, the generating using an inverse kinematic decoder machine-learned model.
A non-transitory computer-readable storage medium storing a set of instructions that, when executed by one or more computer processors, causes the one or more computer processors to perform operations, the operations comprising:
- combining variable numbers and types of supplied inputs into a single input, the variable numbers and types of supplied inputs corresponding to one or more effector constraints for one or more joints of a character;
- transforming the single input into a pose embedding, the pose embedding being a machine-learned representation of the single input;
- and
- expanding the pose embedding into a pose representation output, the pose representation output including local rotation data and global position data for the one or more joints of the character, wherein the expanding of the pose embedding into a pose representation output includes translating the pose embedding into unconstrained predictions of 3D joint positions for the one or more joints, the translating using a global position decoder machine learned model, and generating the local rotation angles for the one or more joints, the generating using an inverse kinematic decoder machine-learned model.
A method comprising:
- combining variable numbers and types of supplied inputs into a single input, the variable numbers and types of supplied inputs corresponding to one or more effector constraints for one or more joints of a character;
- transforming the single input into a pose embedding, the pose embedding being a machine-learned representation of the single input;
- and
- expanding the pose embedding into a pose representation output, the pose representation output including local rotation data and global position data for the one or more joints of the character, wherein the expanding of the pose embedding into a pose representation output includes translating the pose embedding into unconstrained predictions of 3D joint positions for the one or more joints, the translating using a global position decoder machine learned model, and generating the local rotation angles for the one or more joints, the generating using an inverse kinematic decoder machine-learned model.
Dependencies are inferred from the claim text.