Back to results
US 12,731,317 B2Granted
Physics-based image generation using one or more neural networks
AssigneeNVIDIA Corporation
InventorsTingwu Wang, Yunrong Guo, Cheng Xie, Xue Bin Peng, Sanja Fidler
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques are presented to generate images representing realistic motion or activity. In at least one embodiment, one or more neural networks are used to generate one or more images of one or more characters performing one or more actions based, at least in part, upon one or more physical capabilities of the one or more characters.
Claims24
A processor, comprising:
- one or more circuits to:
- cause one or more neural networks to generate one or more actions of one or more characters based, at least in part, on:
- propagating respective information between adjacent nodes of a character graph to update respective information of the adjacent nodes of a given character graph of the one or more characters, wherein the respective information of an individual node of the adjacent nodes for a given character graph of the one or more characters indicates:
- one or more physical capabilities for the one or more characters;
- a current state for the individual node for the given character graph;
- and
- a target state for the individual node for the given character graph;
- and;
- cause a rendering of one or more images depicting a performance of the one or more actions based, at least in part, on the updated respective information of the adjacent nodes of the given character graph, wherein the one or more actions depicted in the one or more images are performed according to the one or more physical capabilities of the one or more characters to perform the one or more actions.
6 more elements
A system comprising one or more processors to:
- cause one or more neural networks to one or more actions of one or more characters based, at least in part, on:
- propagating respective information between adjacent nodes of a character graph to update respective information of the adjacent nodes of a given character graph of the one or more characters, wherein the respective information of an individual node of the adjacent nodes for a given character graph of the one or more characters indicates:
- one or more physical capabilities for the one or more characters;
- a current state for the individual node for the given character graph;
- and
- a target state for the individual node for the given character graph;
- and;
- cause a rendering of one or more images depicting a performance of the one or more actions based, at least in part, on the updated respective information of the adjacent nodes of the given character graph, wherein the one or more actions depicted in the one or more images are performed according to the one or more physical capabilities of the one or more characters to perform the one or more actions.
5 more elements
A method comprising:
- causing one or more neural networks to generate one or more actions of one or more characters based, at least in part, on:
- propagating respective information between adjacent one or more nodes of a character graph to update respective information of the adjacent nodes of a given character graph of the one or more characters, wherein the respective information of an individual node of the adjacent nodes for a given character graph of the one or more characters represents information indicates:
- one or more physical capabilities for the one or more characters;
- a current state for the individual node for the given character graph;
- and
- a target state for the individual node for the given character graph;
- and;
- causing a rendering of one or more images depicting a performance of the one or more actions based, at least in part, on the updated respective information of the adjacent nodes of the given character graph, wherein the one or more actions depicted in the one or more images are performed according to the one or more physical capabilities of the one or more characters to perform the one or more actions.
5 more elements
An image generation system, comprising:
- one or more processors to cause one or more neural networks to generate one or more actions of one or more characters based, at least in part, on:
- propagating respective information between adjacent nodes of a character graph to update respective information of the adjacent nodes of a given character graph of the one or more characters, wherein the respective information of an individual node of the adjacent nodes for a given character graph of the one or more characters indicates:
- one or more physical capabilities for the one or more characters;
- a current state for the individual node for the given character graph;
- and
- a target state for the individual node for the given character graph;
- and;
- cause a rendering of one or more images depicting a performance of the one or more actions based, at least in part, on the updated respective information of the adjacent nodes of the given character graph, wherein the one or more actions depicted in the one or more images are performed according to the one or more physical capabilities of the one or more characters to perform the one or more actions.
5 more elements
Dependencies are inferred from the claim text.