Object image completion
AssigneeNVIDIA Corporation
InventorsSanja Fidler, David Acuna Marrero, Seung Wook Kim, Karsten Julian Kreis, Huan Ling
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques to generate complete depictions of objects based on a partial depiction of the object. In at least one embodiment, an image of a complete object is generated by one or more neural networks, based on an image of a portion of the object, using an encoder of the one or more neural networks trained using training data generated from output of a decoder of the one or more neural networks.
Claims29
A processor, comprising:
- one or more circuits to use one or more neural networks to generate an image of a complete object based, at least in part, on an image of a portion of the object, wherein an encoder of the one or more neural networks is trained using training data generated based, at least in part, on output of a decoder of the one or more neural networks.
A system, comprising:
- one or more processors to train one or more neural networks to generate an image of a complete object based, at least in part, on an image of a portion of the object, wherein an encoder of the one or more neural networks is trained using training data generated based, at least in part, on output of a decoder of the one or more neural networks.
The system of 9, wherein parameters of the decoder are frozen while training the encoder using the training data generated based, at least in part, on output of the decoder.
A method, comprising:
- training a neural network to generate an image of a complete object based, at least in part, on an image of a portion of the object, wherein an encoder of the one or more neural networks is trained using training data generated based, at least in part, on output of a decoder of the one or more neural networks.
A machine-readable medium having stored thereon instructions, which if performed by one or more processors, cause the one or more processors to at least:
- alter an image to incorporate a depiction of a complete object that is generated, by one or more neural networks, from a depiction of a portion of the object in the image, the one or more neural networks trained using training data generated based, at least in part, on output of a decoder of the one or more neural networks.
Dependencies are inferred from the claim text.