Unlock instant, AI-driven research and patent intelligence for your innovation.

Portrait recognition method, portrait recognition device, and terminal device

A portrait recognition and portrait technology, applied in the field of portrait recognition, can solve problems affecting the accuracy of face detection and achieve the effect of accurate second portrait area

Active Publication Date: 2018-12-28
GUANGDONG OPPO MOBILE TELECOMM CORP LTD
View PDF14 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of this, the embodiment of the present application provides a portrait recognition method to solve the problem in the prior art that affects the accuracy of face detection when there are multiple faces in the screen

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Portrait recognition method, portrait recognition device, and terminal device
  • Portrait recognition method, portrait recognition device, and terminal device
  • Portrait recognition method, portrait recognition device, and terminal device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] figure 1 A flowchart of a portrait recognition method provided in Embodiment 1 of the present application is shown. In the present application, the portrait threshold is automatically adjusted according to the number of identified portraits, and then the portraits are re-identified according to the adjusted portrait threshold, so that the recognition result can be improved. The accuracy of , is detailed as follows:

[0042] Step S11, identifying whether the preview screen includes the first portrait;

[0043] Specifically, after the camera preview mode of the terminal device (such as a mobile phone, a tablet computer, a digital camera, etc.) is turned on, the preview interface will display a preview image of an object that can be photographed by the lens.

[0044] In this step, whether the preview image includes a human portrait is identified through a face recognition algorithm or a convolutional neural network, if so, step S12 is performed; otherwise, step S11 is per...

Embodiment 2

[0058] Figure 4 A flowchart of another portrait recognition method provided in Embodiment 2 of the present application is shown. In this embodiment, Step S43 and Step S44 are the refinement steps of Step S13 in Embodiment 1. Step S41, Step S42, Step S44 Step S45 and step S46 are respectively the same as step S11 , step S12 , step S14 , and step S15 in the first embodiment, and will not be repeated here.

[0059] Step S41, identifying whether the preview screen includes the first portrait;

[0060] Step S42, if the preview picture includes the first portrait, count the number of the first portrait included in the preview picture;

[0061] Step S43, compare the counted number of the first portrait with the number of portraits in a preset comparison table, the preset comparison table stores the corresponding relationship between the number of different portraits and the portrait threshold, and the number of portraits Inversely proportional to the portrait threshold;

[0062] ...

Embodiment 3

[0089] Corresponding to Embodiment 1 and Embodiment 2, Figure 5 A schematic structural diagram of a human portrait recognition device provided in Embodiment 3 of the present application is shown, and the human portrait identification device can be applied to a terminal device. For convenience of description, only parts related to this embodiment are shown.

[0090] The portrait identification device includes: a first portrait identification unit 51 , a first portrait count unit 52 , a portrait threshold adjustment unit 53 , a second portrait identification unit 54 , and a second portrait frame selection unit 55 . in:

[0091] The first portrait recognition unit 51 is used to identify whether the preview screen includes the first portrait;

[0092] Specifically, whether the preview image includes a human portrait is identified through a face recognition algorithm or a convolutional neural network.

[0093] The number counting unit 52 of the first portrait is configured to co...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present application is applicable to the technical field of portrait recognition, and provides a portrait recognition method, a portrait recognition device and a terminal device, the method comprising: identifying whether a preview screen includes a first portrait; if the preview screen includes a first portrait, counting the number of the first portrait included in the preview screen; adjusting a portrait threshold according to the number of the first portrait; re-identifying the portrait included in the preview screen according to the adjusted portrait threshold value, and setting the re-identified portrait as a second portrait; boxing to select the area where the second portrait is located. The above method can ensure that the selected portrait area is more accurate.

Description

technical field [0001] The present application belongs to the technical field of portrait recognition, and in particular, relates to a portrait identification method, a portrait identification device, a terminal device, and a computer-readable storage medium. Background technique [0002] At present, when shooting portraits, in order to obtain better shooting effects, for example, in order to obtain better focusing effects, smart terminals (such as mobile phones) will automatically detect faces, and select the area where faces are detected. After performing the photographing action, the smart terminal will automatically focus on the area selected by the frame, so that the photographed face is clearer. [0003] In the existing face detection method, by comparing the detected feature with a preset face feature, if the similarity between the two is greater than a preset similarity threshold, the feature is determined to be a face feature. However, in actual situations, when th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00H04N5/232
CPCG06V40/172H04N23/611H04N23/667
Inventor 张弓
Owner GUANGDONG OPPO MOBILE TELECOMM CORP LTD