Back to results
US 12,592,010 B2Granted
Neural network-based image lighting
AssigneeNVIDIA Corporation
InventorsTing-Chun Wang, Ming-Yu Liu, Koki Nagano, Sameh Khamis, Jan Kautz
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques are presented to generate image data. In at least one embodiment, one or more neural networks are used to cause a lighting effect to be applied to one or more objects within one or more images based, at least in part, on synthetically generated images of the one or more objects.
Claims30
A processor, comprising:
- one or more circuits to use one or more neural networks to cause a lighting effect to be applied to one or more objects within one or more images, wherein a first portion of the one or more neural networks is updated during a first stage based, at least in part, on synthetically generated images of the one or more objects, and a second portion of the one more neural networks that corresponds to predicting surface reflectivity is updated during a second stage based, at least in part, on captured images of one or more physical objects while keeping the first portion of the one or more neural networks fixed.
A system, comprising:
- one or more processors to use one or more neural networks to cause a lighting effect to be applied to one or more objects within one or more images, wherein a first portion of the one or more neural networks is updated during a first stage based, at least in part, on synthetically generated images of the one or more objects, and a second portion of the one more neural networks that corresponds to predicting surface reflectivity is updated during a second stage based, at least in part, on captured images of one or more physical objects while keeping the first portion of the one or more neural networks fixed.
A method, comprising:
- using one or more neural networks to cause a lighting effect to be applied to one or more objects within one or more images, wherein a first portion of the one or more neural networks is updated during a first stage based, at least in part, on synthetically generated images of the one or more objects, and a second portion of the one more neural networks that corresponds to predicting surface reflectivity is updated during a second stage based, at least in part, on captured images of one or more physical objects while keeping the first portion of the one or more neural networks fixed.
A non-transitory machine-readable medium having stored thereon a set of instructions, which if performed by one or more processors, cause the one or more processors to at least:
- use one or more neural networks to cause a lighting effect to be applied to one or more objects within one or more images, wherein a first portion of the one or more neural networks is updated during a first stage based, at least in part, on synthetically generated images of the one or more objects, and a second portion of the one more neural networks that corresponds to predicting surface reflectivity is updated during a second stage based, at least in part, on captured images of one or more physical objects while keeping the first portion of the one or more neural networks fixed.
An image generation system, comprising:
- one or more processors to use one or more neural networks to cause a lighting effect to be applied to one or more objects within one or more images, wherein a first portion of the one or more neural networks is updated during a first stage based, at least in part, on synthetically generated images of the one or more objects, and a second portion of the one more neural networks that corresponds to predicting surface reflectivity is updated during a second stage based, at least in part, on captured images of one or more physical objects while keeping the first portion of the one or more neural networks fixed;
- and
- memory for storing network parameters for the one or more neural networks.
Dependencies are inferred from the claim text.