Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Low-quality video face recognition method based on SICNN

A face recognition, low-quality technology, applied in the field of face recognition, can solve problems such as low efficiency of low-quality video recognition, and achieve the effect of reducing training and testing time, accurate classification results, and reducing computational complexity

Pending Publication Date: 2021-03-30
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The present invention adopts a low-quality video face recognition method based on SICNN (super-identity convolutional neural network), which can effectively improve the low-efficiency problem of low-quality video recognition and achieve a good recognition effect

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
  • Low-quality video face recognition method based on SICNN
  • Low-quality video face recognition method based on SICNN
  • Low-quality video face recognition method based on SICNN

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] A low-quality video face recognition method based on SICNN, comprising the following steps:

[0033] Step 1, data preprocessing, split the low-quality video data in the data set into image frames, face detection and cropping into 32*40px face images, divide the image set into training set and test set using algorithm, and the data set The size ratio is 7:3. For example, the COX data set can be used, and the training samples and test samples have been divided into ten divisions using the COX data set, and the results are the average of ten experiments.

[0034]The COX face dataset aims to solve the problems of video-to-still (V2S), still-to-video (S2V) and video-to-video (V2V) face recognition. The dataset contains 1,000 subjects, and each subject simulates a video surveillance scene, capturing 1 high-quality still image and 3 video sequences (cam1, cam2, cam3). After face detection and data preprocessing, the number of image frames containing faces in most video seque...

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 low-quality video face recognition method based on an SICNN, and the method comprises the following steps: firstly carrying out the classification of key point positions representing the direction of a face through employing a clustering algorithm according to the position features of the face in a video, and selecting key frames; then, establishing an SICNN reconstruction model, extracting features through a reconstruction network and an identification network, obtaining reconstruction loss and identification loss respectively, defining identity loss accordingly, training the reconstruction network by using an alternate training strategy, and obtaining a frame image with high resolution and more identity features; and finally, inputting the reconstructed frame image into an identification network Inception Resnet v2, extracting depth features for classification and identification, and voting identification results of all image frames to obtain a video identification result. The method is applied to low-quality video face recognition, and the low-quality video face recognition accuracy is effectively improved.

Description

technical field [0001] The technical field of face recognition of the present invention relates in particular to a low-quality video face recognition method based on SICNN. Background technique [0002] In recent years, the continuous development of computer vision has led to more and more technologies landing into practical products in daily life. With the rise of deep neural networks, face recognition technology has developed rapidly. Among them, image face recognition has achieved excellent results, but the research on video face recognition is relatively less than people's expectations. This is because video face recognition not only faces the same lighting, occlusion, posture and other problems as image face recognition, but also the image frame quality of video in practical applications (such as in surveillance scenes) is usually not as good as that of images. Currently, video face recognition methods are divided into two categories: classical methods and deep learni...

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): G06K9/00G06K9/34G06K9/46G06K9/62G06T3/40G06N3/04G06N3/08
CPCG06T3/4053G06N3/08G06V40/161G06V40/168G06V10/267G06V10/44G06N3/045G06F18/213G06F18/214G06F18/2415
Inventor 袁家斌陆要要何珊
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
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