Unlock instant, AI-driven research and patent intelligence for your innovation.

Training method and device of facial recognition model

A face recognition and training method technology, applied in the field of face recognition model training methods and devices, can solve the problem of low recognition accuracy, inability to recognize standard face images and face images with occluders at the same time, and inability to Identification and other issues

Active Publication Date: 2020-11-10
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF8 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the current face recognition technology cannot recognize standard face images and face images with occluders at the same time, and the accuracy of recognizing face images with occluders on the face is low, or even unrecognizable question

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
  • Training method and device of facial recognition model
  • Training method and device of facial recognition model
  • Training method and device of facial recognition model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

[0027] Some face recognition models in the related art do not have the ability to recognize occluded faces, or the recognition rate of occluded faces is very low, so they cannot meet the scene of occluded face recognition. However, some models that have the ability to recognize occluded faces sacrifice the recognition rate of unoccluded standard faces in order to imp...

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 training method and device of a facial recognition model, and relates to the technical field of artificial intelligence, deep learning and computer vision, in particular to the technical field of face recognition. According to the specific implementation scheme, the method comprises the steps that a first training image is acquired, the first training image is an unshielded face image, after multiple shielding object images are acquired, the multiple shielding object images are fused to the unshielded face image to generate multiple second training images, the first training image and the second training images are input into a face recognition model, and a face recognition result is obtained to train a face recognition model. Therefore, the facial recognition model is trained by adopting the unshielded face image and the plurality of second training images obtained by fusion, so that the trained face recognition model can accurately recognize the unshielded face image and the shielded face image at the same time; the technical problems that when an existing facial recognition model is used for recognizing a face image with an obstruction, the accuracy rate is low, and even the face image with the obstruction cannot be recognized are solved.

Description

technical field [0001] This application relates to artificial intelligence, deep learning and computer vision, specifically to the technical field of face recognition, and in particular to a training method and device for a face recognition model. Background technique [0002] At present, face recognition technology has been widely used in video surveillance, security and financial payment and other occasions. Face recognition in actual natural scenes, because the face may be covered by masks, scarves and other occluders in a large area, resulting in the loss of a large number of human facial features. [0003] In existing face recognition technologies, face recognition results can be obtained through accurate recognition based on collected face recognition images. However, the current face recognition technology cannot recognize standard face images and face images with occluders at the same time, and the accuracy of recognizing face images with occluders on the face is lo...

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): G06K9/00G06K9/62
CPCG06V40/166G06V40/171G06F18/214G06V40/172G06V10/774Y02T10/40G06V10/247G06V10/34G06V10/72G06V10/7715G06V40/162G06V10/267G06V10/273
Inventor 范彦文余席宇张刚刘经拓王海峰丁二锐韩钧宇
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD