Image attribute processing system, image attribute processing apparatus, and method of processing image attributes

Patent metadata

Field	Value
Publication number	US-11093974-B2
Application number	US-201916238845-A
Country	US
Kind code	B2
Filing date	Jan 3, 2019
Priority date	Jan 12, 2018
Publication date	Aug 17, 2021
Grant date	Aug 17, 2021

Abstract

An information processing system includes circuitry configured to store, in a memory, one or more feature value patterns associated with appearance attribute of one or more groups of persons calculated from a plurality of acquired image data using machine learning, in which each one of the groups assumed to have a unique group value being different for each one of the groups, receive image data of a target person input as analysis target data, analyze an appearance attribute of the target person in the image data using the one or more feature value patterns associated with the appearance attribute of the one or more groups stored in the memory, and output a response corresponding to an analysis result of the appearance attribute of the target person.

First claim

What is claimed is:

1. An information processing system comprising:
circuitry configured to
store, in a memory, one or more feature value patterns associated with an appearance attribute of at least one of clothing or accessories of one or more groups of persons calculated from a plurality of acquired image data using machine learning, in which each one of the groups assumed to have a unique group value being different for each one of the groups;
receive image data of a target person input as analysis target data;
analyze an appearance attribute of the target person in the image data using the one or more feature value patterns associated with the appearance attribute of the one or more groups stored in the memory; and
output a response corresponding to an analysis result of the appearance attribute of the target person, wherein the circuitry is further configured to
receive a designation of an area set for the image data input as the analysis target data, and a designation of particular appearance attributes specified in the set area in advance, the set area including one or more set areas, and
output a response in accordance with a comparison result of the appearance attribute of the target person and the particular appearance attribute designated for the seat area.

2. The information processing system of claim 1 ,
wherein the circuitry calculates the one or more feature value patterns associated with the appearance attribute of the one or more groups of persons from the plurality of acquired image data using machine learning based on an assumption that a group value of each one of the groups is the same value in each one of the groups, and stores the one or more feature value patterns associated with the appearance attribute of the one or more groups of persons in the memory.

3. The information processing system of claim 1 , further comprising:
an image capture device that captures an image of the target person to output the image data of the target person, wherein the circuitry receives the image data of the target person from the image capturing device.

4. The information processing system according to claim 1 ,
wherein the one or more groups of persons include at least a first group having a first group value and a second group having a second group value different from the first group value of the first group,
wherein when the circuitry receives the plurality of acquired image data as image data of the first group having the first group value, the circuitry applies one or more processing uniquely set for the appearance attribute of the first group to duplicate an amount of the image data of the first group, and when the circuitry receives the plurality of acquired image data as image data of the second group having the second group value, the circuitry applies one or more processing uniquely set for the appearance attribute of the second group to duplicate an amount of the image data of the second group.

5. The information processing system of claim 1 ,
wherein the circuitry analyzes and outputs a matching level of the appearance attribute of the target person in the image data, input as the analysis target data, and the appearance attribute of the one or more groups stored in the memory.

6. The information processing system of claim 1 ,
wherein the circuitry calculates a correlation value between the appearance attributes of the groups using the one or more feature value patterns associated with the appearance attribute of the one or more groups,
wherein the circuitry analyzes a correlation level of the appearance attribute of the target person in the image data and the appearance attribute of each of the groups and outputs a response in accordance with the correlation level of the appearance attribute of the target person in the image data and the appearance attribute of each of the groups.

7. The information processing system of claim 1 ,
wherein the circuitry uses advertisement definition data defining advertisement information in accordance with the appearance attribute of the target person to output specific advertisement information in accordance with the appearance attribute of the target person in the image data, input and analyzed as the analysis target data.

8. The information processing system of claim 1 ,
wherein the circuitry adds the appearance attribute of the target person to existing information of the target person.

9. The information processing system of claim 8 ,
wherein based on information related to a change of the appearance attribute of a plurality of persons accumulated for a given period of time, the circuitry predicts an over-time change of the appearance attribute of the target person in the image data, input and analyzed as the analysis target data, and outputs a prediction result of the over-time change of the appearance attribute of the target person.

10. The information processing system of claim 9 ,
wherein the circuitry predicts a life event of the target person based on the information related to the change of the appearance attribute of the plurality of persons.

11. An information processing apparatus comprising:
circuitry configured to
store, in a memory, one or more feature value patterns associated with an appearance attribute of at least one of clothing or accessories of one or more groups of persons calculated from a plurality of acquired image data using machine learning, in which each one of the groups assumed to have a unique group value being different for each one of the groups;
receive image data of a target person input as analysis target data;
analyze an appearance attribute of the target person in the image data using the one or more feature value patterns associated with the appearance attribute of the one or more groups stored in the memory; and
output a response corresponding to an analysis result of the appearance attribute of the target person, wherein the circuitry is further configured to
receive a designation of an area set for the image data input as the analysis target data, and a designation of particular appearance attributes specified in the set area in advance, the set area including one or more set areas, and
output a response in accordance with a comparison result of the appearance attribute of the target person and the particular appearance attribute designated for the seat area.

12. A method of processing information, the method comprising:
receiving image data of a target person input as analysis target data;
receiving a designation of an area set for the image data input as the analysis target data, and a designation of particular appearance attributes specified in the set area in advance, the set area including one or more set areas,
analyzing an appearance attribute of a target person in image data input as analysis target data using one or more feature value patterns associated with appearance attribute of at least one of clothing or accessories of one or more groups of persons calculated from a plurality of acquired image data using machine learning and stored in a memory, in which each one of the groups assumed to have a unique group value being different for each one of the groups; and
outputting a response corresponding to an analysis result of the appearance attribute of the target person; and
outputting a response in accordance with a comparison result of the appearance attribute of the target person and the particular appearance attribute designated for the set area.

Assignees

• Tazuke Tomoyuki
• Hamabe Muneyuki
• Kakinuma Akihiro
• Ricoh Co Ltd

Inventors

• Tazuke Tomoyuki
• Hamabe Muneyuki
• Kakinuma Akihiro

Classifications

• G06V30/19173

  Classification techniques · CPC title

• G06F18/241

  relating to the classification model, e.g. parametric or non-parametric approaches · CPC title

• G06F18/2148

  characterised by the process organisation or structure, e.g. boosting cascade · CPC title

• G06F18/211

  Selection of the most significant subset of features · CPC title

• G06V30/244

  using graphical properties, e.g. alphabet type or font · CPC title