Method and device for recommending cloth coordinating information

Patent metadata

Field	Value
Publication number	US-11263473-B2
Application number	US-201916579465-A
Country	US
Kind code	B2
Filing date	Sep 23, 2019
Priority date	Aug 16, 2019
Publication date	Mar 1, 2022
Grant date	Mar 1, 2022

Abstract

Disclosed are a method and an apparatus for recommending clothing coordination information in the 5G communication environment by executing an artificial intelligence (AI) algorithm and/or a machine learning algorithm, which have been mounted therein. A method for recommending clothing coordination information according to an embodiment of the present disclosure may include classifying style information through an analysis of a collected clothing image, calculating a harmony matching score between clothing images included in the classified style information or a TPO matching score on a time, a place, and an occasion of each of the clothing images and registering the calculated score in a database, recognizing a clothing coordination request speech voice received from a user, and confirming the speech intention of the clothing coordination request speech voice of the user, and recommending clothing coordination information corresponding to the speech intention, based on the harmony matching score or the TPO matching score.

First claim

What is claimed is:

1. A method for recommending clothing coordination information, the method comprising:
classifying style information through an analysis of a collected clothing image;
calculating a harmony matching score between clothing images comprised in the classified style information or a TPO (time, place and occasion) matching score on a time, a place, and an occasion of each of the clothing images and registering the calculated score in a database;
recognizing a clothing coordination request speech voice received from a user, and confirming the speech intention of the clothing coordination request speech voice of the user; and
recommending clothing coordination information corresponding to the speech intention, based on the harmony matching score or the TPO matching score registered in the database,
wherein the recommending comprises:
recommending the clothing coordination information in the order of the highest score combination of the harmony matching scores of the clothing images registered in the database, based on the speech intention of the clothing coordination request speech voice.

2. A method for recommending clothing coordination information, the method comprising:
classifying style information through an analysis of a collected clothing image;
calculating a harmony matching score between clothing images comprised in the classified style information or a TPO (time, place and occasion) matching score on a time, a place, and an occasion of each of the clothing images and registering the calculated score in a database;
recognizing a clothing coordination request speech voice received from a user, and confirming the speech intention of the clothing coordination request speech voice of the user; and
recommending clothing coordination information corresponding to the speech intention, based on the harmony matching score or the TPO matching score registered in the database,
wherein the classifying the style information comprises:
classifying appearance style information of the collected clothing image by analyzing a feature vector extracted from the collected clothing image by using a first deep neural network model previously trained so as to classify appearance style information of a clothing image by analyzing a feature vector extracted from the clothing image; and
classifying abstract style information of the collected clothing image by analyzing the appearance style information generated from the collected clothing image by using a second deep neural network model previously trained so as to classify abstract style information of a clothing image by analyzing the appearance style information of the clothing image.

3. The method claim 1 , wherein the confirming the speech intention comprises:
extracting information on at least one of a time, a place, or an occasion to wear the coordinated clothing from the clothing coordination request speech voice received from the user, and
wherein the recommending the clothing coordination information comprises recommending the clothing coordination information corresponding to the speech intention based on the extracted information on at least one of the time, the place, or the occasion and the TPO matching score of each of the clothing images stored in the database.

4. The method of claim 1 , wherein the registering comprises:
calculating the harmony matching score of the collected clothing image on a clothing image owned by the user and registering the harmony matching score in the database, based on the similarity between style information comprising the appearance style information and the abstract style information on the collected clothing image, and style information on the clothing image owned by the user registered in the database.

5. The method of claim 1 , wherein the registering comprises:
calculating the harmony matching score of the collected clothing image on an image for which clothing coordination has been completed and registering the harmony matching score in the database, based on the similarity between the collected clothing image, and the image for which the clothing coordination has been completed and which has been collected from clothing shopping malls.

6. A non-transitory computer readable recording medium in which a computer program for executing the method of claim 1 by using a computer has been stored.

7. The method of claim 1 , further comprising collecting weather information of a date when the clothing coordination request speech voice is received,
wherein the recommending comprises:
recommending the clothing coordination information in the order of the highest score combination of the harmony matching scores of the clothing images registered in the database, based on the speech intention of the clothing coordination request speech voice and weather information of the date when the clothing coordination request speech voice is received.

8. The method of claim 1 , further comprising:
collecting a clothing image that the user intends to purchase from a user terminal;
calculating the harmony matching score between the clothing image that the user intends to purchase and the clothing image registered in the database by analyzing the clothing image that the user intends to purchase and style information of the clothing image that the user intends to purchase by using a third deep neural network model previously trained so as to calculate harmony matching score between clothing images through analysis of the clothing image and the style information; and
transmitting purchase recommendation information or purchase holding information to the user terminal based on the harmony matching score.

9. An apparatus for recommending clothing coordination information, the apparatus comprising:
a processor; and
a memory coupled to the processor,
the processor configured to:
classify style information through an analysis of a collected clothing image;
calculate a harmony matching score between clothing images comprised in the classified style information or a TPO (time, place and occasion) matching score on a time, a place, and an occasion of each of the clothing images and registering the calculated score in the database;
recognize a clothing coordination request speech voice received from a user, and confirming the speech intention of the clothing coordination request speech voice of the user; and
recommend clothing coordination information corresponding to the speech intention, based on the harmony matching score or the TPO matching score registered in the database,
wherein the processor is further configured to:
recommend the clothing coordination information in the order of the highest score combination of the harmony matching scores of the clothing images registered in the database, based on the speech intention of the clothing coordination request speech voice.

10. The apparatus of claim 9 , wherein the processor is further configured to:
collect the clothing image that the user intends to purchase from a user terminal,
calculate the harmony matching score between the clothing image that the user intends to purchase and the clothing image registered in the database by analyzing the clothing image that the user intends to purchase and style information of the clothing image that the user intends to purchase by using a third deep neural network model previously trained so as to calculate harmony matching score between clothing images through analysis of the clothing images and the style information, and
transmit purchase recommendation information or purchase holding information to the user terminal based on the harmony matching score.

11. The apparatus of claim 9 , wherein the pro

Assignees

• Lg Electronics Inc

Inventors

• Kim Ji Hye
• Park Jin Sung
• Jin Moon Sub
• Huh Ji Young
• Kim Beom Oh

Classifications

• G06F16/55 Primary

  Clustering; Classification · CPC title

• G06V10/462

  Salient features, e.g. scale invariant feature transforms [SIFT] · CPC title

• G06V10/764

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

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title