Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Face aligning method and device

A face alignment and face technology, applied in the field of face recognition, can solve the problems of large memory space and large size of the system model.

Active Publication Date: 2018-05-22
BEIJING SOHU NEW MEDIA INFORMATION TECH
View PDF15 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of this, the embodiment of the present invention provides a method and device for face alignment, which solves the problem of large system model size and large memory space 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
  • Face aligning method and device
  • Face aligning method and device
  • Face aligning method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] refer to figure 1 , which shows a schematic flowchart of a method for calculating an interpolation matrix provided by an embodiment of the present invention. In this embodiment, the method includes:

[0052] S101: Determine all feature points of the face;

[0053]In this embodiment, technicians can set all feature points used for face alignment according to actual needs, for example, 128 feature points or 68 feature points can be used.

[0054] For example: the feature point is the coordinate point that can be used to form the shape of the face, such as figure 2 as shown, figure 2 The points on can be represented as feature points.

[0055] S102: Divide the process of face alignment into multiple stages;

[0056] S103: Determine the feature points and regression series of each stage; wherein, the number of feature points in the latter stage is greater than the number of feature points in the previous stage, and the number of feature points in the last stage is equ...

Embodiment 2

[0069] refer to image 3 , which shows a schematic flowchart of a method for face alignment provided by an embodiment of the present invention. In this embodiment, the method includes:

[0070] S201: Update the shape of the face according to the position coordinates of the feature points and the extracted feature values ​​of the face in the current stage; the current stage is any one of all stages in the face alignment process, the The feature points are the coordinate points that constitute the shape of the face;

[0071] In this embodiment, it can be known from the introduction of Embodiment 1 that the technician divides the process of face alignment into multiple stages, wherein the multiple stages are at least two stages, and each stage includes a different number of feature points. The number of feature points in a stage is greater than the number of feature points in the previous stage, and the number of feature points in the last stage is equal to the number of all pre...

Embodiment approach 1

[0082] If the current regression level is not the first regression level of the first stage, determine the position coordinates of all the feature points according to the position coordinates of the feature points in the current stage and the second interpolation matrix;

[0083] According to the calculated position coordinates of all the feature points, the face reference shape of the current regression level in the current stage is determined.

[0084] Among them, the second interpolation matrix indicates that it is used to multiply the position coordinates of the feature points in the current stage to obtain the estimated values ​​of the position coordinates of all feature points. For example, suppose the face alignment process is divided into 3 stages. If the current When the stage is the first stage, all the feature points of the last stage can be obtained through interpolation of the position coordinates of the feature points of the first stage; if the current stage is th...

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 aligning method and device. The method comprises the following steps of updating the face shape according to the position coordinate of the feature point and the extracted face feature values at the current stage, wherein the current stage refers to any one stage in all stages in the face aligning execution process; the feature point is a coordinate point forming theface shape; under the condition that the current stage is not the last stage in the all stages, determining the position coordinate of the feature point in a next stage according to a preset first interpolation matrix and the position coordinate of the feature point of the current stage; cyclically executing the steps until the current stage is the last stage in all stages; outputting the updatedface shape at the current stage. Therefore on the premise of not damaging the system model performance, the system model volume is reduced, the calculating speed is accelerated.

Description

technical field [0001] The invention relates to the field of face recognition, in particular to a method and device for face alignment. Background technique [0002] With the development of technology in the field of computer vision, the alignment technology for facial feature points is becoming more and more perfect. At present, it usually includes two types of algorithms: the first type is based on traditional machine learning algorithms, such as boosting, random forest, etc.; the second type is based on The algorithm of the neural network, the most typical convolutional neural network, can achieve better accuracy based on the convolutional neural network algorithm, but it is slower in practical applications and difficult to deploy. Therefore, in engineering applications, traditional machine learning methods occupy an important position. At present, most of the traditional machine learning algorithms obtain the positioning results of the feature points through step-by-ste...

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): G06K9/00
CPCG06V40/1347
Inventor 赵壁原肖锋
Owner BEIJING SOHU NEW MEDIA INFORMATION TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com