Object recognition method and apparatus

Patent metadata

Field	Value
Publication number	US-12033369-B2
Application number	US-202217668101-A
Country	US
Kind code	B2
Filing date	Feb 9, 2022
Priority date	Aug 9, 2019
Publication date	Jul 9, 2024
Grant date	Jul 9, 2024

Abstract

A method for optimizing a photographing pose of a user, where the method is applied to an electronic device, and the method includes: displaying a photographing interface of a camera of the electronic device; obtaining a to-be-taken image in the photographing interface; determining, based on the to-be-taken image, that the photographing interface includes a portrait; entering a pose recommendation mode; and presenting a recommended human pose picture to a user in a predetermined preview manner, where the human pose picture is at least one picture that is selected from a picture library through metric learning and that has a top-ranked similarity to the to-be-taken image, and where the similarity is an overall similarity obtained by fusing a background similarity and a foreground similarity.

First claim

What is claimed is:

1. A method for recommending a similar human pose picture, the method comprising:
receiving an input picture, wherein the input picture comprises a first portrait;
selecting, as a recommended human pose picture, at least one picture that has a highest similarity to the input picture from a picture library through metric learning that is based on a multi-level environmental information feature, wherein the recommended human pose picture comprises a second portrait, and wherein the multi-level environmental information feature comprises a scene feature, an object spatial distribution feature, and a foreground human feature; and
presenting the recommended human pose picture in a predetermined preview manner.

2. The method of claim 1 , further comprising receiving a recommendation preference setting of a user, wherein selecting the at least one picture comprises selecting, as the recommended human pose picture, the at least one picture based on a recommendation preference of the user and through metric learning that is based on the multi-level environmental information feature, and wherein the recommended human pose picture meets the recommendation preference setting of the user.

3. The method of claim 1 , wherein selecting the at least one picture comprises:
performing feature extraction processing on the input picture to obtain a first feature of the input picture;
calculating, through metric learning based on the multi-level environmental information feature, a similarity between the first feature and a second feature that is of each image and that is in a feature library, wherein the feature library is obtained by extracting a predetermined quantity of dimensions of features from each picture in the picture library; and
selecting at least one recommended picture corresponding to a top-ranked similarity as the recommended human pose picture from the picture library based on a calculation result of the similarity between the first feature and the second feature.

4. The method of claim 1 , further comprising:
receiving a recommendation preference setting of a user; and
screening human pose pictures in the picture library to obtain, as a final recommended human pose picture, a picture that meets the recommendation preference setting of the user.

5. The method of claim 1 , wherein the first portrait is a photographing object, wherein receiving the input picture comprises receiving a plurality of input pictures that are at different angles and that comprise the photographing object, and wherein selecting the at least one picture comprises:
calculating, through metric learning that is based on the multi-level environmental information feature, a picture that is in the picture library and that is most similar to each of the input pictures;
ranking all most similar pictures in the picture library; and
selecting, from the most similar pictures, at least one top-ranked picture as the recommended human pose picture.

6. The method of claim 1 , further comprising:
receiving a user-defined picture uploaded by a user; and
updating the picture library to include the user-defined picture.

7. An apparatus for recommending a picture, comprising:
a memory configured to store a computer program; and
a processor coupled to the memory and configured to execute the computer program to:
receive an input picture, wherein the input picture comprises a first portrait;
select, as a recommended human pose picture, at least one picture that has a highest similarity to the input picture from a picture library through metric learning that is based on a multi-level environmental information feature, wherein the recommended human pose picture comprises a second portrait, and wherein the multi-level environmental information feature comprises a scene feature, an object spatial distribution feature, and a foreground human feature; and
present the recommended human pose picture in a predetermined preview manner.

8. The apparatus of claim 7 , wherein the processor is further configured to receive a recommendation preference setting of a user, wherein the processor is configured to select the at least one picture by selecting, as the recommended human pose picture, the at least one picture based on a recommendation preference of the user and through metric learning that is based on the multi-level environmental information feature, and wherein the recommended human pose picture meets the recommendation preference setting of the user.

9. The apparatus of claim 7 , wherein the processor is configured to select the at least one picture by:
performing feature extraction processing on the input picture to obtain a first feature of the input picture;
calculating, through metric learning that is based on the multi-level environmental information feature, a similarity between the first feature and a second feature that is of each image and that is in a feature library, wherein the feature library is obtained by extracting a predetermined quantity of dimensions of features from each picture in the picture library; and
selecting at least one recommended picture corresponding to a top-ranked similarity as the recommended human pose picture from the picture library based on a calculation result of the similarity between the first feature and the second feature.

10. The apparatus of claim 7 , wherein the processor is further configured to:
receive a recommendation preference setting of the user; and
screen human pose pictures in the picture library to obtain, as a final recommended human pose picture, a picture that meets the recommendation preference setting of the user.

11. The apparatus of claim 7 , wherein the first portrait is a photographing object, wherein the processor is configured to receive the input picture by receiving a plurality of input pictures that are at different angles and that comprise the photographing object, and wherein the processor is configured select the at least one picture by:
calculating, through metric learning that is based on the multi-level environmental information feature, a picture that is in the picture library and that is most similar to each of the input pictures;
ranking all most similar pictures in the picture library; and
selecting, from the most similar pictures, at least one top-ranked picture as the recommended human pose picture.

12. The apparatus of claim 7 , wherein the processor is further configured to:
receive a user-defined picture uploaded by a user; and
update the picture library to include the user-defined picture.

13. A non-transitory computer storage medium configured to store instructions that, when executed by a processor of an electronic device, cause the electronic device to perform a method for recommending a similar human pose picture, the method comprising:
receiving an input picture, wherein the input picture comprises a first portrait;
selecting, as a recommended human pose picture, at least one picture that has a highest similarity to the input picture from a picture library through metric learning that is based on a multi-level environmental information feature, wherein the recommended human pose picture comprises a second portrait, and wherein the multi-level environmental information feature comprises a scene feature, an object spatial distribution feature, and a foreground human feature; and
presenting the recommended human pose picture in a predetermined preview manner.

14. The non-transitory computer storage medium of claim 13 , wherein the method further comprises receiving a recommendation preference setting of a user, wherein selecting the at least one

Assignees

• Huawei Tech Co Ltd

Inventors

• Huang Fei
• Wu Xiaofei
• Li Zhihao
• Xu Songcen
• Liu Jianzhuang
• Yan Youliang
• Qian Li
• Huang Xueyan

Classifications

• G06V10/40

  Extraction of image or video features · CPC title

• G06V40/103

  Static body considered as a whole, e.g. static pedestrian or occupant recognition · CPC title

• G06V10/454

  Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN] · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/08

  Learning methods · CPC title