US 11,379,692 B2 · Learning method, storage medium and image processing device

Abstract

According to one embodiment, a learning method for causing a second statistical model to learn using a first statistical model is provided. The method includes obtaining a first learning image, cutting out each local area of the obtained first learning image, and obtaining a first prediction value output from the first statistical model by inputting each local area to the first statistical model and obtaining a second prediction value output from the second statistical model by inputting the entire area of the first learning image to the second statistical model, and causing the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

Claims15

Dependent claims

A learning method for causing a second statistical model for outputting a prediction value corresponding to an entire area of an image based on input of the entire area of the image to learn using a first statistical model for outputting a prediction value corresponding to a local area of the image based on input of the local area of the image, the method comprising:

obtaining a first learning image;
cutting out each local area of the obtained first learning image, and obtaining a first prediction value corresponding to each local area output from the first statistical model by inputting each local area to the first statistical model;
obtaining a second prediction value corresponding to an entire area of the obtained first learning image output from the second statistical model by inputting the entire area of the first learning image to the second statistical model;
and
causing the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

2 more elements

A non-transitory computer-readable storage medium having stored thereon a computer program which is executable by a computer and causes a second statistical model for outputting a prediction value corresponding to an entire area of an image based on input of the entire area of the image to learn using a first statistical model for outputting a prediction value corresponding to a local area of the image based on input of the local area of the image, the computer program comprising instructions capable of causing the computer to execute functions of:

obtaining a first learning image;
cutting out each local area of the obtained first learning image, and obtaining a first prediction value corresponding to each local area output from the first statistical model by inputting each local area to the first statistical model;
obtaining a second prediction value corresponding to an entire area of the obtained first learning image output from the second statistical model by inputting the entire area of the first learning image to the second statistical model;
and
causing the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

2 more elements

An image processing device for causing a second statistical model for outputting a prediction value corresponding to an entire area of an image based on input of the entire area of the image to learn using a first statistical model for outputting a prediction value corresponding to a local area of the image based on input of the local area of the image, the device comprising:

a hardware processor is configured to:
obtain a first learning image;
cut out each local area of the obtained first learning image and obtains a first prediction value corresponding to each local area output from the first statistical model by inputting each local area to the first statistical model;
obtain a second prediction value corresponding to an entire area of the obtained first learning image output from the second statistical model by inputting the entire area of the first learning image to the second statistical model;
and
cause the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

3 more elements

Dependencies are inferred from the claim text.

Facts

Application

16/808,586Filed 4 Mar 2020

Priority

JP JP2019-11200817 Jun 2019

Published

US 2020/0394464 A117 Dec 2020

Granted

5 Jul 2022

Term ends

Not estimatedThe source records don’t support an estimate.

Classification19 codes

G06N 3/08
Neural networks: learning methods
G06F 17/18
Complex mathematical operations: for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression…
G06F 18/2148
Generating training patterns; Bootstrap methods, e.g. bagging or boosting: characterised by the process organisation or structure, e.g. boosting…
G06F 18/2193
Validation; Performance evaluation; Active pattern learning techniques: based on specific statistical tests

Abstract

Claims15

Dependent claims

obtaining a first learning image;
cutting out each local area of the obtained first learning image, and obtaining a first prediction value corresponding to each local area output from the first statistical model by inputting each local area to the first statistical model;
obtaining a second prediction value corresponding to an entire area of the obtained first learning image output from the second statistical model by inputting the entire area of the first learning image to the second statistical model;
and
causing the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

2 more elements

obtaining a first learning image;
cutting out each local area of the obtained first learning image, and obtaining a first prediction value corresponding to each local area output from the first statistical model by inputting each local area to the first statistical model;
obtaining a second prediction value corresponding to an entire area of the obtained first learning image output from the second statistical model by inputting the entire area of the first learning image to the second statistical model;
and
causing the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

2 more elements

a hardware processor is configured to:
obtain a first learning image;
cut out each local area of the obtained first learning image and obtains a first prediction value corresponding to each local area output from the first statistical model by inputting each local area to the first statistical model;
obtain a second prediction value corresponding to an entire area of the obtained first learning image output from the second statistical model by inputting the entire area of the first learning image to the second statistical model;
and
cause the second statistical model to learn based on a difference between the first prediction value and the second prediction value.

3 more elements

Dependencies are inferred from the claim text.

Learning method, storage medium and image processing device

Life of this patent

Abstract

Claims15

Learning method, storage medium and image processing device

Life of this patent

Abstract

Claims15