Label-dependent loss function for discrete ordered regression model

Patent metadata

Field	Value
Publication number	US-2023410478-A1
Application number	US-202118035635-A
Country	US
Kind code	A1
Filing date	Nov 8, 2021
Priority date	Nov 9, 2020
Publication date	Dec 21, 2023
Grant date	—

Abstract

A processing apparatus is provided that is configured to perform operations including obtaining a plurality of images having been evaluated by different sources such that each source has classified each of the plurality of image as being a member of one of a predefined class, generating a distribution array identifying a number of times each image of the plurality of images has been classified into each of the predefined classes, generating, for each predefined class, a loss function based on the ratio of a number of images in other classes of the predefined classes to a number of images to this predefined classes, providing the generated loss function for each predefined class as evaluation parameters to a model, and using the generated loss function to determine that the model classifies raw image data as being a member of one of the predefined classes according to a predetermined accuracy threshold.

First claim

We claim:

1 . A method comprising
obtaining a plurality of images having been evaluated by different sources such that each source has classified each of the plurality of image as being a member of one of a predefined class; generating a distribution array identifying a number of times each image of the plurality of images has been classified into each of the predefined classes; generating, for each predefined class, a loss function based on the ratio of a number of images in other classes of the predefined classes to a number of images to this predefined classes; providing the generated loss function for each predefined class as evaluation parameters to a model; using the generated loss function to determine that the model classifies raw image data as being a member of one of the predefined classes according to a predetermined accuracy threshold.

2 . The method according to claim 1 , wherein obtaining a plurality of images includes
obtaining a plurality of labeled image sets wherein,
each image set of the plurality of image sets includes common images,
each of the plurality of images in each image set is labeled as being in a particular class selected from a predefined set of classes; and
each image set has been labeled by an evaluator; and each image could be labeled differently by a different evaluator

3 . The method according to claim 1 , further comprising
generating likelihood sets of images corresponding to each of the predefined classes wherein each likelihood set includes all images in the particular class as being classified from the different sources.

4 . The method according to claim 1 , wherein the predefined classes represent a degree characterizing an image feature.

5 . The method according to claim 1 , wherein the predefined classes hold some uncertainty due to human perception.

6 . The method according to claim 1 , wherein the predefined class is sharpness

7 . The method according to claim 1 , wherein each of the predefined classes represent a different degree of image sharpness by human perception.

8 . The method according to claim 1 , wherein each of the predefined classes represents a metric of human perception and the generated loss function makes the classified raw image data to match the human perception.

9 . The method according to claim 1 , further comprising
modifying at least one parameter of the model other than the generated loss function; and using the updated model with the generated loss function to determine whether the updated model classifies raw image data according to the predetermined accuracy threshold.

10 . A processing apparatus comprising:
one or more memories storing instructions; one or more processors that, upon executing the stored instructions; are configured to perform operations including
obtaining a plurality of images having been evaluated by different sources such that each source has classified each of the plurality of image as being a member of one of a predefined class;
generating a distribution array identifying a number of times each image of the plurality of images has been classified into each of the predefined classes;
generating, for each predefined class, a loss function based on the ratio of a number of images in other classes of the predefined classes to a number of images to this predefined classes;
providing the generated loss function for each predefined class as evaluation parameters to a model;
using the generated loss function to determine that the model classifies raw image data as being a member of one of the predefined classes according to a predetermined accuracy threshold.

11 . The processing apparatus according claim 1 , wherein the obtained plurality of images include a plurality of labeled image sets wherein,
each image set of the plurality of image sets includes common images, each of the plurality of images in each image set is labeled as being in a particular class selected from a predefined set of classes; and each image set has been labeled by an evaluator; and each image could be labeled differently by a different evaluator.

12 . The processing apparatus according to claim 10 , wherein execution of the stored instructions further configures the one or more processors to perform operations including
generating likelihood sets of images corresponding to each of the predefined classes wherein each likelihood set includes all images in the particular class as being classified from the different sources.

13 . The processing apparatus according to claim 10 , wherein the predefined classes represent a degree characterizing an image feature.

14 . The processing apparatus according to claim 10 , wherein the predefined classes hold some uncertainty due to human perception.

15 . The processing apparatus according to claim 10 , wherein the predefined class is sharpness

16 . The processing apparatus according to claim 10 , wherein each of the predefined classes represent a different degree of image sharpness by human perception.

17 . The processing apparatus according to claim 10 , wherein each of the predefined classes represents a metric of human perception and the generated loss function makes the classified raw image data to match the human perception.

18 . The processing apparatus according to claim 1 , wherein execution of the stored instructions further configures the one or more processors to perform operations including
modifying at least one parameter of the model other than the generated loss function; and using the updated model with the generated loss function to determine whether the updated model classifies raw image data according to the predetermined accuracy threshold.

Assignees

• Canon Usa Inc

Inventors

• Cao Xiwu

Classifications

• G06V10/766 Primary

  using regression, e.g. by projecting features on hyperplanes · CPC title

• G06V10/776

  Validation; Performance evaluation · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06T7/00 Primary

  Image analysis · CPC title

• G06T2207/20076

  Probabilistic image processing · CPC title