Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Neck key point detection method and device and detection model training method and device

A technology for detecting models and key points, applied in the field of image processing, can solve problems such as improper connection between the face and neck, and achieve the effect of improving robustness and accuracy

Active Publication Date: 2022-05-27
北京新氧科技有限公司
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003]However, in the prior art, the main focus of the portrait stitching operation is the detection of key points of the face, and various stitching and deformation operations are performed according to the detected key points of the face , but did not pay attention to the influence of the neck area on the connection between the face and the neck, resulting in the inappropriate connection between the face and the neck after subsequent related operations on the face

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
  • Neck key point detection method and device and detection model training method and device
  • Neck key point detection method and device and detection model training method and device
  • Neck key point detection method and device and detection model training method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] figure 1 A flowchart of an embodiment of a method for training a neck key point detection model according to an exemplary embodiment of the present invention, as shown in figure 1 As shown, the training method of the neck key point detection model includes the following steps:

[0058] Step 101: Determine the actual slope of the straight line determined by the marked neck key point and the reference point in the sample image.

[0059] Among them, the neck is usually a key organ connecting the face and the human body, and there is a certain geometric constraint relationship between it and the human body, that is, the two sides of the neck are horizontally symmetrical with respect to the center line of the neck. Therefore, by marking a key point on the left and right sides of the neck, you can Determines the position of the neck relative to the face.

[0060] In a possible implementation manner, after acquiring the data set, for each sample image in the data set, locate...

Embodiment 2

[0090] image 3 This is an embodiment flowchart of a neck key point detection method according to an exemplary embodiment of the present invention, based on the above figure 1 Based on the illustrated embodiment, as image 3 As shown, the neck key point detection method includes the following steps:

[0091] Step 301: Preprocess the received user image to obtain an image to be detected.

[0092] The preprocessing of the user image is to process the user image into an image suitable for prediction by the input model.

[0093] In a possible implementation manner, by performing a preset type of data enhancement processing on the user image, and determining the proportion of the area of ​​the face area including the neck in the processed user image to the processed user image, if the proportion exceeds The preset ratio means that the proportion of the face area including the neck in the image is relatively high, and the processed user image can be directly determined as the image...

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 invention discloses a neck key point detection method and device and a detection model training method and device. The training method comprises the following steps: determining an actual slope of a straight line determined by a marked neck key point and a reference point in a sample graph; inputting the sample graph into a pre-constructed neck key point detection model, and learning and outputting neck key points by the neck key point detection model; calculating a loss value by using the neck key points output by the model, the marked neck key points and the actual slope; and when the loss value is greater than a preset value, optimizing network parameters of the neck key point detection model according to the loss value, and continuing training until the loss value is lower than the preset value. According to the method, the actual slope between the marked neck key point and the reference point is obtained, loss calculation is carried out by using the actual slope, the prediction result and the marking result in the training stage, and network parameter optimization is carried out according to the calculation result, so that the robustness and accuracy of model detection are improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a neck key point detection method and device, and a detection model training method and device. Background technique [0002] At present, there are more and more applications of image processing for portraits. In the field of medical beauty, the stitching operation technology for portraits is the most widely used simulation plastic surgery. [0003] However, in the prior art, the image stitching operation mainly focuses on the detection of key points of the face, and performs various stitching and deformation operations according to the detected key points of the face, but does not pay attention to the influence of the neck area on the connection between the face and the neck. As a result, the connection between the face and the neck is inappropriate after the subsequent operations are performed on the face. SUMMARY OF THE INVENTION [0004] The purpose of the present...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06V40/10G06V10/774G06K9/62
CPCG06F18/214Y02T10/40
Inventor 张译友杨浩杰张梦洁
Owner 北京新氧科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products