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

Tiny-face recognition method based on generative adversarial networks

A face recognition and network technology, applied in the field of face recognition, can solve the problems of not capturing tiny faces and the decrease of detection rate, so as to achieve the effect of improving the recognition rate and promoting the improvement of recognition accuracy

Active Publication Date: 2018-07-27
HARBIN INST OF TECH
View PDF5 Cites 41 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to solve the problem that the deep learning technology at the present stage does not have the ability to capture tiny faces in complex backgrounds, and when fa

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
  • Tiny-face recognition method based on generative adversarial networks
  • Tiny-face recognition method based on generative adversarial networks
  • Tiny-face recognition method based on generative adversarial networks

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0028] Specific implementation mode one: the miniature face recognition method based on generation confrontation network of this embodiment, such as Figure 9 shown, including:

[0029]Step 1: Establish a training database. For example, the WIDER FACE database can be used as the training database, or the face image size in the WIDER FACE database is between 10 and 30 pixels to construct the training database. Hard tiny face detection problem. This embodiment also supports users to build databases by collecting images of real scenes. The tiny human face referred to in the present invention is a human face image whose size is between 10 and 30 pixels.

[0030] Step 2, use the face detector to predict the face position of each picture in the training database, and intercept the first high-resolution face image and the first high-resolution non-face image; and process the first high-resolution A low-resolution face image and a low-resolution non-face image are obtained from 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 provides a tiny-face recognition method based on generative adversarial networks. The method is provided for solving the shortcomings that current-stage face-detection technology cannotcapture tiny faces under complex backgrounds, and can cause serious decreasing of detection rates when face detection is carried out on the basis of distorted images, and includes: using an existing face detector to predict a face position of each picture in a training database, and intercepting and saving real-face and non-face images; carrying out downsampling according to face images and non-face pictures to obtain corresponding low-resolution images; constructing the generative adversarial networks, wherein the generative adversarial networks include a generator and a discriminator; usinghigh-resolution face and non-face images and the corresponding low-resolution face and non-face images to train the generative adversarial networks; and marking a position of a face in an input picture according to scores of the discriminator on face candidate areas obtained from the existing face detector. The method is suitable for use in recognition and detection of faces.

Description

technical field [0001] The present invention relates to the field of face recognition, in particular to a small face recognition method based on a generative confrontation network. Background technique [0002] Face detection is a very important basic research topic in the field of machine vision. It is the basic technology for advanced tasks such as face parsing, face verification, and face tagging. In addition, face recognition is playing an increasingly important role in the fields of friend recommendation, automatic tagging of photo albums, security and anti-terrorism. It also provides a good solution for modern identification and has a wide range of application prospects. [0003] Due to the important theoretical research value and urgent practical application requirements of face detection technology, the corresponding technology for face detection is also constantly developing and updating, which can be roughly divided into two categories: traditional methods of face...

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): G06K9/00G06K9/62G06T3/40G06N3/04G06N3/08
CPCG06N3/08G06T3/4007G06T3/4076G06T2207/20084G06T2207/20081G06T2207/30201G06V40/161G06V40/172G06N3/045G06F18/214
Inventor 张永强丁明理白延成李贤杨光磊董娜
Owner HARBIN INST OF 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