US 11,688,109 B2 · Generating differentiable procedural materials

Abstract

The present disclosure relates to using end-to-end differentiable pipeline for optimizing parameters of a base procedural material to generate a procedural material corresponding to a target physical material. For example, the disclosed systems can receive a digital image of a target physical material. In response, the disclosed systems can retrieve a differentiable procedural material for use as a base procedural material in response. The disclosed systems can compare a digital image of the base procedural material with the digital image of the target physical material using a loss function, such as a style loss function that compares visual appearance. Based on the determined loss, the disclosed systems can modify the parameters of the base procedural material to determine procedural material parameters for the target physical material. The disclosed systems can generate a procedural material corresponding to the base procedural material using the determined procedural material parameters.

Claims20

Dependent claims

A non-transitory computer-readable medium storing instructions thereon that, when executed by at least one processor, cause the at least one processor to perform operations comprising:

retrieving, from a database, a procedural material;
identifying a plurality of nodes of the procedural material;
converting nodes of the plurality of nodes into differentiable nodes;
and
generating a differentiable procedural material with the differentiable nodes.

2 more elements

A system comprising:

at least one memory device comprising a base procedural material;
and
at least one processor coupled to the at least one memory device that causes the system to perform operations comprising:
identifying a plurality of nodes of the base procedural material;
converting nodes of the plurality of nodes into differentiable nodes;
generating a differentiable base procedural material from the differentiable nodes;
receiving a digital image of a target physical material;
extracting a first set of deep neural features from the digital image of the target physical material;
extracting a second set of deep neural features from a digital image of the base procedural material;
modifying differentiable nodes of the base procedural material via back propagation based on comparing the first set of deep neural features and the second set of deep neural features;
and
generating a target procedural material corresponding to the target physical material using the modified differentiable nodes.

9 more elements

Dependencies are inferred from the claim text.

Generating differentiable procedural materials

Life of this patent

Generating differentiable procedural materials

Life of this patent

Abstract

Claims20