Particle-based inverse kinematic rendering system

A computer-implemented method comprising:

• as implemented by a user computing device configured with specific computer-executable instructions for executing a game application,
• generating a virtual game environment;
• determining a first pose of a first virtual character model within the virtual game environment, the first virtual character model comprising a plurality of rigid bodies connected by a plurality of joints, each joint connected to at least one rigid body, wherein the first pose of the first virtual character model is a first arrangement of the plurality of rigid bodies and the plurality of joints of the first virtual character model;
• rendering at least a portion of the first pose of the first virtual character model within a frame during runtime of the game application;
• receiving an instruction to change the first pose of the first virtual character model in response to a first movement of at least one joint of the first virtual character model;
• determining a second pose for the first virtual character model by:
• calculating an estimated pose of the first virtual character model based at least in part on a first prediction model of the plurality of prediction models wherein the estimated pose is not output for rendering during runtime of the game application;
• and
• using the estimated pose as an input to a second prediction model of the plurality of prediction models;
• calculating the second pose of the first virtual character model based, at least in part, on an output of the second prediction model of the plurality of prediction models, wherein the estimated pose and the second pose are calculated in response to the first movement of the at least one joint of the first virtual character model, wherein the estimated pose is a pose distinct from the first post and the second pose;
• and
• rendering at least a portion of the second pose of the first virtual character model within a frame during runtime of the game application.

A non-transitory computer-readable storage medium having stored thereon computer-executable instructions that, when executed, direct a computing system to perform a method comprising:

• executing a game application;
• generating a virtual game environment;
• determining a first pose of a first virtual character model within the virtual game environment, the first virtual character model comprising a plurality of rigid bodies connected by a plurality of joints, each joint connected to at least one rigid body, wherein the first pose of the first virtual character model is a first arrangement of the plurality of rigid bodies and the plurality of joints of the first virtual character model;
• rendering at least a portion of the first pose of the first virtual character model within a frame during runtime of the game application;
• receiving an instruction to change the first pose of the first virtual character model in response to a first movement of at least one joint of the first virtual character model;
• determining a second pose for the first virtual character model by:
• calculating an estimated pose of the first virtual character model based at least in part on a first prediction model of the plurality of prediction models, wherein the estimated pose is not output for rendering during runtime of the game application;
• and
• using the estimated pose as an input to a second prediction model of the plurality of prediction models;
• calculating the second pose of the first virtual character model based, at least in part, on an output of the second prediction model of the plurality of prediction models, wherein the estimated pose and the second pose are calculated in response to the first movement of the at least one joint of the first virtual character model, wherein the estimated pose is a pose distinct from the first post and the second pose;
• rendering at least a portion of the second pose of the first virtual character model within a frame during runtime of the game application.