Supercharge Your Innovation With Domain-Expert AI Agents!

Image beautification processing method and system based on Retinaface algorithm, storage medium and equipment

A processing method and image technology, applied in the field of image processing, can solve the problem of low speed of face key point detection

Active Publication Date: 2022-04-29
江西中业智能科技有限公司
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Based on this, the object of the present invention is to provide an image beauty treatment method, system, storage medium and equipment based on the Retinaface algorithm, which is used to solve the problem of low detection speed of human face key points in the prior art

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
  • Image beautification processing method and system based on Retinaface algorithm, storage medium and equipment
  • Image beautification processing method and system based on Retinaface algorithm, storage medium and equipment
  • Image beautification processing method and system based on Retinaface algorithm, storage medium and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] see figure 1 , shows the image beauty processing method based on the Retinaface algorithm in the first embodiment of the present invention, the method includes steps S101 to S103:

[0044] S101. Acquire the original image data, extract multiple data features of the original image data through the backbone network in the algorithm, and synthesize the multiple data features into a feature pyramid.

[0045] Specifically, multiple data features of the original image data are extracted through the backbone network in the Retinaface algorithm. The backbone network includes multiple sub-networks, and each sub-network extracts data features of different dimensions. The sub-networks include classification sub-networks, and face frame detection. sub-network and face key points sub-network.

[0046] S102. Perform feature detection through each sub-network on the data features after the synthetic feature pyramid, obtain multiple trained sub-networks, merge multiple trained sub-net...

Embodiment 2

[0050] Please check figure 2 , shows the image beauty processing method based on the Retinaface algorithm in the second embodiment of the present invention, the method includes steps S201 to S203:

[0051] S201. Acquire the original image data, extract multiple data features of the original image data through the backbone network in the algorithm, and synthesize the multiple data features into a feature pyramid.

[0052] Specifically, the backbone network includes multiple sub-networks, and each sub-network extracts data features of different dimensions. The sub-networks include a classification sub-network, a face frame detection sub-network, and a face key point sub-network.

[0053] Obtain training data pictures and coordinates from the original image data and scale them to 640*640 pixels, and input them into the backbone network. Specifically, use MobilieNetV3 (0.25 times) as the backbone network of Retinaface, for example and not limitation, in other embodiments In , ot...

Embodiment 3

[0090] see Figure 5 , showing the image beauty treatment system based on the Retinaface algorithm in the third embodiment of the present invention, the system includes:

[0091] The acquisition module is used to acquire the original image data, extract a plurality of data features of the original image data through the backbone network in the Retinaface algorithm, and synthesize a plurality of the data features into a feature pyramid, the backbone network includes a plurality of sub-networks, Each of the sub-networks extracts data features of different dimensions, and the sub-networks include a classification sub-network, a face frame detection sub-network and a face key point sub-network;

[0092] The training module is used to perform feature detection on the data features after the synthetic feature pyramid through each of the sub-networks to obtain a plurality of trained sub-networks, and merge a plurality of the trained sub-networks to obtain a trained backbone network ...

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 provides an image beautification processing method and system based on a Retinaface algorithm, a storage medium and equipment, and the method comprises the steps: obtaining original image data, extracting a plurality of data features of the original image data, synthesizing the data features into a feature pyramid, and carrying out the feature detection through each sub-network, thereby obtaining a plurality of trained sub-networks, combining the trained sub-networks to obtain a trained backbone network, and establishing a target detection model; and obtaining face data and performing data preprocessing, obtaining a plurality of face key point coordinates from the face data after data preprocessing through the target detection model, and obtaining a beautified image through the face key points. According to the image beautification processing method and system based on the Retinaface algorithm, the storage medium and the equipment, key point detection can be directly carried out through the trained model, the detection speed is improved, and the problem that in the prior art, the face key point detection speed is low is solved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image beautification processing method, system, storage medium and equipment based on Retinaface algorithm. Background technique [0002] With the improvement of the computing power of mobile devices and the development of neural network model optimization, it is possible to perform face beautification and face thinning locally on the mobile terminal in real time. [0003] Beautification refers to the use of image processing technology to beautify the portraits in images or videos to better meet the aesthetic needs of users. In the field of live broadcast, it often involves the use of mobile phone cameras to take selfies or shoot other people's scenes. In these scenes, real-time beauty is widely demanded, including filters, face-lifting, and eye-enlarging functions. Among them, face-lifting and eye-enlarging functions depend on the key to the human face. poin...

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): G06T5/00
CPCG06T2207/10004G06T2207/20081G06T2207/20084G06T2207/30201G06T5/77
Inventor 史绍阳李介刘丹张恒星高园岗杨忠
Owner 江西中业智能科技有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com