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

Deformed face detection based on multidirectional fusion attention

A kind of attention and multi-directional technology, applied in the field of deformation face detection technology, can solve the problems of poor robustness, high error rate, large number of parameters, etc., and achieve the effect of enhanced representation and reliable detection

Pending Publication Date: 2022-06-10
CHANGSHA UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Its purpose is to solve the problems of high error rate, poor robustness, and large number of parameters in previous methods.

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
  • Deformed face detection based on multidirectional fusion attention
  • Deformed face detection based on multidirectional fusion attention
  • Deformed face detection based on multidirectional fusion attention

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments of the description. As shown in Figure 1, a

[0021] Each step is described in detail below.

[0022] In step A1, in the face deformation attack, the face region is usually located in the center of the image. In order to accurately draw the

[0024]

[0025]

[0026] In step A3, a multi-directional fusion attention module is designed. The specific steps are: given an input X, first

[0027]

[0029]

[0030] The above two transformations aggregate features along two spatial directions to obtain a pair of direction-aware feature maps. by letter

[0032] In the formula, [.,.] is the concatenate operation along the spatial dimension, δ is the nonlinear activation function, and f is the pair of spatial information.

[0033] g

[0034] g

[0035] Here σ is the sigmoid activation function. In order to reduce the complexity and computational overhead o...

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 a deformed face detection method based on multidirectional fusion attention for face deformation detection, and the method comprises the following steps: 1) segmenting and normalizing the face of an image according to eye coordinates detected by a dlib mark point detector; 2) considering the position information ignored by channel attention, and proposing a new attention module; and 3) a double-branch convolutional network is fused, so that the detection precision is improved. And (4) classifying the final feature map by using an SVM (Support Vector Machine).

Description

Deformed face detection based on multi-directional blended attention technical field The present invention relates to face fusion attack detection field, especially a kind of deformed face based on multi-directional fusion attention Detection Technology. Background technique [0002] Face recognition technology has made great achievements in the field of security. But over the past few years, researchers have pointed out various potential deficiencies of biometric systems. Recently, for face and The vulnerability of fingerprinting has been established. Morphing techniques can be used to create artificial biometric samples that are Biometric information similar to two (or more) data subjects in the Image and Feature domains. If it contains deformed individual characteristic information image or template is penetrated into the biometric system, the subjects making up the deformed image will be classified according to a single registration template. Successfully v...

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): G06V40/16G06V40/40G06V10/44G06V10/764G06V10/80G06V10/82G06N3/04G06K9/62
CPCG06N3/048G06N3/045G06F18/2411G06F18/253
Inventor 彭烨凡龙敏徐启航
Owner CHANGSHA UNIVERSITY OF SCIENCE AND TECHNOLOGY
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