Joint twist generation for animation
AssigneeElectronic Arts Inc.
InventorsWolfram Sebastian Starke, Harold Henry Chaput, Yiwei Zhao
Life of this patent
Estimating the term…
Abstract
The present disclosure provides embodiments for joint twist generation for animation. The system can utilize a neural network, also referred to as a deep neural network, which utilizes machine learning processes in order to create animation data that are more life-like and realistic. The system can obtain a set of axis vectors for a rig of a virtual character model; obtain a twist model for the rig; input the set of axis vectors to the twist model to obtain a set of twist vectors; and determine animation data based on the set of axis vectors and the set of twist vectors.
Claims20
A system comprising:
- at least one memory storing instructions and game application data;
- and
- at least one processor configured to execute the instructions to perform operations, the operations including:
- execute an instance of a game application based in part on the game application data;
- generate a virtual game environment within the game application;
- determine a first pose of a first virtual character model within the virtual game environment, the first virtual character model comprising a virtual rig including a plurality of a rigid bodies connected by a plurality of joints, wherein the first pose is a first arrangement of the rigid bodies and joints of the virtual rig;
- receive an instruction to change the first pose of the first virtual character model based at least in part on the movement of a joint of the first virtual character model;
- obtain a set of axis vectors for the virtual rig of the first virtual character model;
- input the set of axis vectors to a twist model to generate a set of twist vectors;
- determine animation data based at least in part on the set of axis vectors and the set of twist vectors;
- determine an updated pose of first virtual character model based, at least in part, on the animation data;
- render the first virtual character model based, at least in part on the updated pose of the first virtual character model;
- and
- generate instructions to output a frame comprising at least a portion of the updated pose of the first virtual character model for a user display.
A computer-implemented method comprising:
- executing an instance of a game application based in part on the game application data;
- generating a virtual game environment within the game application;
- determining a first pose of a first virtual character model within the virtual game environment, the first virtual character model comprising a virtual rig including a plurality of a rigid bodies connected by a plurality of joints, wherein the first pose is a first arrangement of the rigid bodies and joints of the virtual rig;
- receiving an instruction to change the first pose of the first virtual character model based at least in part on the movement of a joint of the first virtual character model;
- obtaining a set of axis vectors for the virtual rig of the first virtual character model;
- inputting the set of axis vectors to a twist model to generate a set of twist vectors;
- determining animation data based at least in part on the set of axis vectors and the set of twist vectors;
- determining an updated pose of first virtual character model based, at least in part, on the animation data;
- rendering the first virtual character model based, at least in part on the updated pose of the first virtual character model;
- and
- generating instructions to output a frame comprising at least a portion of the updated pose of the first virtual character model for a user display.
A non-transitory computer-readable storage medium having stored thereon instructions that, when executed, direct a computing system to perform a method, the method comprising:
- executing an instance of a game application based in part on the game application data;
- generating a virtual game environment within the game application;
- determining a first pose of a first virtual character model within the virtual game environment, the first virtual character model comprising a virtual rig including a plurality of a rigid bodies connected by a plurality of joints, wherein the first pose is a first arrangement of the rigid bodies and joints of the virtual rig;
- receiving an instruction to change the first pose of the first virtual character model based at least in part on the movement of a joint of the first virtual character model;
- obtaining a set of axis vectors for the virtual rig of the first virtual character model;
- inputting the set of axis vectors to a twist model to generate a set of twist vectors;
- determining animation data based at least in part on the set of axis vectors and the set of twist vectors;
- determining an updated pose of first virtual character model based, at least in part, on the animation data;
- rendering the first virtual character model based, at least in part on the updated pose of the first virtual character model;
- and
- generating instructions to output a frame comprising at least a portion of the updated pose of the first virtual character model for a user display.
Dependencies are inferred from the claim text.