Information processing apparatus and information processing method to generate image processing algorithm based on machine learning process
AssigneeSony Corporation
InventorHiroshi Arai
Life of this patent
Estimating the term…
Abstract
An information processing apparatus includes a control unit that creates a plurality of learning information items including an input image and a teacher image as an expected value by image-processing the input image in accordance with a scenario described with a program code, and supplies the created plurality of learning information items to a machine learning module that composes an image processing algorithm by machine learning.
Claims6
An information processing apparatus, comprising:
- at least one processor configured to:
- generate a plurality of learning information items for an original image, wherein
- each learning information item of the plurality of learning information items includes an input image and a teacher image,
- a frame rate of the teacher image is double a frame rate of the input image,
- the teacher image corresponds to an expected value, based on an image-processing operation on the input image,
- the image-processing operation is based on a scenario associated with the original image, and
- a first learning information item of the plurality of learning information items is different from a second learning information item of the plurality of learning information items based on the scenario;
- and
- supply the plurality of learning information items to a machine learning processor, wherein the machine learning processor generates an image processing algorithm based on a machine learning process of the plurality of learning information items.
An information processing method, comprising:
- in an information processing apparatus:
- generating, by at least one processor of the information processing apparatus, a plurality of learning information items for an original image, wherein
- each learning information item of the plurality of learning information items includes an input image and a teacher image,
- a frame rate of the teacher image is double a frame rate of the input image,
- the teacher image corresponds to an expected value, based on an image-processing operation on the input image,
- the image-processing operation is based on a scenario associated with the original image, and
- a first learning information item of the plurality of learning information items is different from a second learning information item of the plurality of learning information items based on the scenario;
- and
- supplying, by the at least one processor, the plurality of learning information items to a machine learning processor, wherein the machine learning processor generates an image processing algorithm based on a machine learning process of the plurality of learning information items.
A non-transitory computer-readable medium having stored thereon computer-executable instructions, which when executed by an information processing device, cause the information processing device to execute operations, the operations comprising:
- generating a plurality of learning information items for an original image, wherein
- each learning information item of the plurality of learning information items includes an input image and a teacher image,
- a frame rate of the teacher image is double a frame rate of the input image,
- the teacher image corresponds to an expected value, based on an image-processing operation on the input image,
- the image-processing operation is based on a scenario associated with the original image, and
- a first learning information item of the plurality of learning information items is different from a second learning information item of the plurality of learning information items based on the scenario;
- and
- supplying the plurality of learning information items to a machine learning processor, wherein the machine learning processor generates an image processing algorithm based on a machine learning process of the plurality of learning information items.
Dependencies are inferred from the claim text.