Intelligent photographing method and system, and related apparatus

Patent metadata

Field	Value
Publication number	US-11470246-B2
Application number	US-201817284906-A
Country	US
Kind code	B2
Filing date	Oct 15, 2018
Priority date	Oct 15, 2018
Publication date	Oct 11, 2022
Grant date	Oct 11, 2022

Abstract

An intelligent photographing method includes extracting, by a terminal, one or more first tags from common data of a user, where the common data represents an identity feature of the user, extracting, by the terminal, one or more second tags from photographing-related data of the user, where the photographing-related data represents a photographing preference of the user, determining, by the terminal, a third tag based on the one or more first tags and the one or more second tags, and obtaining a picture and adjusting, by the terminal, picture quality of the picture based on a picture quality effect parameter set corresponding to the third tag.

First claim

What is claimed is:

1. An intelligent photographing method implemented by a terminal, wherein the method comprises:
extracting, based on a first mapping relationship, one or more first tags from common data of a user, wherein the common data represents an identity feature of the user, and wherein the first mapping relationship comprises mappings between a plurality of groups of the common data and a plurality of the one or more first tags;
extracting one or more second tags from photographing-related data of the user, wherein the photographing-related data represents a photographing preference of the user;
determining, based on the one or more first tags and the one or more second tags, a third tag;
taking a picture; and
adjusting, based on a picture quality effect parameter set corresponding to the third tag, a picture quality of the picture.

2. The intelligent photographing method of claim 1 , further comprising:
extracting, from the photographing-related data, one or more first photographing-related parameter sets;
inputting the one or more first photographing-related parameter sets to a first neural network model to obtain one or more first score vector sets, wherein the one or more first score vector sets comprise first scores of a plurality of fourth tags, and wherein each of the first scores represents a matching degree between a corresponding first photographing-related parameter set and a corresponding fourth tag; and
determining, based on the first score vector sets, the one or more second tags in the fourth tags.

3. The intelligent photographing method of claim 2 , wherein the one or more second tags comprise one or more fourth tags having first scores that are greater than a first threshold.

4. The intelligent photographing method of claim 2 , wherein the one or more second tags comprise one or more fourth tags having first scores that are highest in a first score vector set corresponding to each of the one or more first photographing-related parameter sets.

5. The intelligent photographing method of claim 2 , wherein before inputting the one or more first photographing-related parameter sets to the first neural network model, the intelligent photographing method further comprises:
obtaining sample data comprising a plurality of groups of first training sets, wherein each of the groups of the first training sets comprises one group of second photographing-related parameter sets and one group of second score vector sets; and
training the first neural network model based on the sample data using a deep learning algorithm.

6. The intelligent photographing method of claim 5 , further comprising:
displaying a first interface comprising a plurality of sample pictures, wherein each of the sample pictures corresponds to the one group of the second photographing-related parameter sets and the one group of the second score vector sets, wherein each of the second photographing-related parameter sets represents picture qualities of the sample pictures, and wherein each of the second score vector sets comprises first scores of a second plurality of fourth tags corresponding to the sample pictures;
receiving, from the user, a first input operation of selecting one or more training pictures from the sample pictures; and
setting, in response to the first input operation, the one or more first photographing-related parameter sets and the second score vector sets that are corresponding to the one or more training pictures as the sample data.

7. The intelligent photographing method of claim 6 , further comprising:
determining whether a quantity of the sample pictures is less than a training quantity; and
extracting one or more groups of the first training sets from a prestored training set database as the sample data when the quantity of the sample pictures is less than the training quantity.

8. The intelligent photographing method of claim 1 , wherein each of the one or more first tags and each of the one or more second tags jointly correspond to an association score, wherein a value of the association score represents a degree of association between each of the one or more first tags and each of the one or more second tags, and wherein the intelligent photographing method further comprises:
determining a total association score of the each of the one or more second tags based on the one or more first tags and the one or more second tags using a formula:

T
i

=

L
1

*

(

∑

k
=
1

R

⁢

W
k

)

+

L
2

,

wherein T i is a total association score of an i th second tag in the one or more second tags, wherein L 1 is a weight of the one or more first tags, wherein L 2 is a weight of the one or more second tags, wherein W k is an association score corresponding to a k th first tag in the one or more first tags and the i th second tag, and wherein R is a quantity of the one or more first tags; and
determining, based on the total association score, the third tag, wherein the third tag is with a highest total association score in the one or more second tags.

9. An electronic device comprising:
a memory configured to store instructions; and
a processor coupled to the memory, wherein the instructions cause the processor to be configured to:
extract, based on a first mapping relationship, extract one or more first tags from common data of a user, wherein the common data represents an identity feature of the user, and wherein the first mapping relationship comprises mapping between a plurality of groups of the common data and a plurality of the one or more first tags;
extract one or more second tags from photographing-related data of the user, wherein the photographing-related data represents a photographing preference of the user;
determine, based on the one or more first tags and the one or more second tags, a third tag;
take a picture; and
adjust, based on a picture quality effect parameter set corresponding to the third tag, a picture quality of the picture.

10. The electronic device of claim 9 , wherein the instructions further cause the electronic device to:
extract, from the photographing-related data, one or more first photographing-related parameter sets;
input the one or more first photographing-related parameter sets to a first neural network model to obtain one or more first score vector sets, wherein the one or more first score vector sets comprise first scores of a plurality of fourth tags, and wherein each of the first scores represents a matching degree between a corresponding first photographing-related parameter set and a corresponding fourth tag; and
determine, based on the first score vector sets, the one or more second tags in the fourth tags.

Assignees

• Huawei Tech Co Ltd

Inventors

• Kang Fengxia
• Chen Shaojun
• Chen Hao
• Zhou Shengfeng

Classifications

• G06V10/82

  using neural networks · CPC title

• G06N3/08 Primary

  Learning methods · CPC title

• G06V10/764

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

• H04N23/64

  Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image · CPC title

• H04N23/617

  Upgrading or updating of programs or applications for camera control · CPC title