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

Face alignment method based on cascade position regression of random forests

A random forest, face alignment technology, applied in computer parts, instruments, character and pattern recognition, etc., can solve the problems of inability to cope with changes in face pose, partial occlusion of the face, and poor robustness of the regressor.

Active Publication Date: 2016-06-01
GUILIN UNIV OF ELECTRONIC TECH
View PDF5 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, the robustness of the regressor obtained by this method is poor, and it cannot cope with the situation where there are large changes in face pose and partial occlusion of 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
  • Face alignment method based on cascade position regression of random forests
  • Face alignment method based on cascade position regression of random forests
  • Face alignment method based on cascade position regression of random forests

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0050] see figure 1 , a method for face alignment based on random forest cascade position regression, including the following steps:

[0051] 1) Get the normalized face picture: read the pictures in the training set image library and the corresponding face attributes, and normalize the pictures. The face attributes include the rectangular area information of the face position, that is, x 1 axis, y 1 Axis, w width, h height information and known key point coordinate information of the calibration is x 2 axis, y 2 Axis information;

[0052] 2) Calculate the average shape of the face: determine 20 initial shapes for each face training sample, except for its own shape, that is, form 810×20 training samples, and rotate and scale the key point coordinate information of the training samples to be similar Transform to calculate the average shape of a face:

[0053] M S = 1 N ( Σ ...

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 alignment method based on cascade position regression of random forests. The face alignment method is characterized in that 1, a normalized face image is acquired; 2, the average shape of the faces can be calculated; 3, a candidate feature point of a face alignment frame is generated; 4, a face shape index grey level is generated; 5, a face shape index feature X is generated; 6, a face alignment frame is built; 7, the face shape is initialized, and after the continuous iteration, the final estimated face shape can be output. The good robustness can be kept under the conditions of the changing illumination, the changing expressions, and the shielding, and the precision can be improved, and the failure rate can be reduced.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, in particular to a random forest-based cascade position regression method for face alignment. Background technique [0002] Face alignment plays a very important role in face recognition, face tracking and 3D face reconstruction, attracting more and more researchers. However, due to the diversity of facial expressions, different lighting conditions and occlusions, there are also It brings difficulties and challenges to the research. [0003] Face alignment algorithms are roughly divided into two categories, one is the optimal face alignment algorithm, and the other is the regression-based face alignment algorithm. [0004] The optimal face alignment algorithm achieves the goal of face alignment by optimizing the error. Its performance depends on the design of the error equation itself and its optimization effect, which is difficult to guarantee. For example, the face alignment ...

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): G06K9/00
CPCG06V40/16G06V40/168G06V40/172
Inventor 莫建文彭倜张彤袁华陈利霞首照宇欧阳宁高宇匡勇建
Owner GUILIN UNIV OF ELECTRONIC 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