Face image processing method based on multi-task learning

A face image, multi-task model technology, applied in the direction of instruments, character and pattern recognition, computer parts, etc., can solve the problems of network structure redundancy and processing time, and achieve the effect of reducing the number of times and saving network reasoning.

Active Publication Date: 2021-11-12
北京万里红科技有限公司
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are obvious shortcomings in this processing method, that is, for face detection tasks, face key point detection tasks, and face tracking tasks, at least three network models need to be trained
Correspondingly, three kinds of data set labels need to be prepared in advance, which are used to train these three network models respectively; and, in the process of network reasoning, three basic networks need to be used for face feature extraction, and the network structure is redundant and processing time consuming

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
  • Face image processing method based on multi-task learning
  • Face image processing method based on multi-task learning
  • Face image processing method based on multi-task learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0034] Aiming at the problems existing in the prior art, the present disclosure provides a face image processing scheme based on multi-task learning. First, the acquired video image frame is input into the face multi-task model for processing to detect the face in the video image frame and output the face position information, face key point position information, and face re-identification (ReID ) forecast information. In one embo...

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 face image processing method based on multi-task learning. The method comprises the steps of: processing a current frame image through a face multi-task model, specifically, outputting face frame position information through a face detection component of the face multi-task model; outputting face key point position information through a face key point extraction component of the face multi-task model; outputting face re-recognition information through a re-recognition component of the face multi-task model; and based on face frame position information and the face re-recognition information, matching with a face in a previous frame to determine a face identifier of a face detected in the current frame. The invention also discloses a method for training and generating the face multi-task model.

Description

technical field [0001] The present disclosure relates to the technical field of image processing, in particular to a face image processing method based on multi-task learning. Background technique [0002] At present, face detection technology, face key point detection technology, and face tracking technology have a wide range of business scopes and application scenarios, but there are almost no methods for end-to-end processing of the three tasks through one model. The existing business solutions basically use the trained face detection model to detect the face in the video image; then, input it into the trained face key point detection model, and output the face key points to Solve the task of facial key point detection. Or, connect the output of the face detection model to the face tracking feature extraction model, and output the face identification (face ID) to solve the face tracking task. In other words, connecting various network models through cascading is the sol...

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): G06K9/00G06K9/62
CPCG06F18/22G06F18/214
Inventor 张小亮王秀贞戚纪纲杨占金其他发明人请求不公开姓名
Owner 北京万里红科技有限公司
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
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap