A method and device for fast detection and training of face images based on pruning model

A face image and training method technology, which is applied in the field of face image detection training, can solve problems such as poor training speed and accuracy, and achieve the effects of ensuring accuracy, reducing volume, and reducing network connections

Active Publication Date: 2021-11-23
YUNNAN UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention is mainly aimed at the fast detection and training method of human face images. In order to solve the problem of poor detection and training speed and accuracy, the present invention provides a fast detection and training method of human face images based on pruning model

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
  • A method and device for fast detection and training of face images based on pruning model
  • A method and device for fast detection and training of face images based on pruning model
  • A method and device for fast detection and training of face images based on pruning model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0051] A fast detection and training method for human face images based on a pruning model, comprising the following steps: establishing a feature extraction model, and performing a pruning operation on the model; using a region candidate network model to generate a candidate frame; determining the content in the candidate frame, Whether it belongs to the foreground or the background; determine the attributes of the target in the candidate frame, and correct the position of the candidate frame containing the target; finally output the image containing the face annotation frame and face attributes. Specifically:

[0052] (1) Step 1: Establish a feature extraction model and perform pruning operations on the model

[0053] The present invention starts with the connection weights of the connections between layers in the feature extraction process, and obtains a model according to the traditional model training method training; starting from the weights of the first layer connectio...

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 method and device for rapid detection and training of human face images based on a pruning model, comprising the following steps: (1) establishing a feature extraction model, and performing a pruning operation on the model; (2) using a region candidate network model , to generate a candidate frame; (3) determine whether the content in the candidate frame belongs to the foreground or the background; (4) determine the attributes of the target in the candidate frame, and calculate the offset of the candidate frame, and remove the candidate frame that is too small or exceeds the boundary , and correct the position of the candidate frame; (5) output the image containing the face annotation frame and face attributes. The present invention is suitable for rapid face detection training, which not only improves the speed of model training, but also ensures the accuracy of the model.

Description

technical field [0001] The invention belongs to the technical field of face image detection and training, in particular to the technical field of a fast detection and training method of human face images based on a pruning model; the invention also designs a device structure for the content of the method. Background technique [0002] At present, face detection technology is frequently used in people's daily life, and its application background has exceeded the scope of face recognition. It has broad application prospects in video detection, content retrieval, digital video processing, etc., among which the most popular A few of the application areas are liveness verification, image search, smart beauty, movie production, etc. In such an application background, the face detection technology has been greatly developed, and the speed and accuracy of the detection model are also continuously improved. Currently, there are two series of algorithms that are extremely prominent. O...

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 Patents(China)
IPC IPC(8): G06K9/00
CPCG06V40/171G06V40/161
Inventor 陶大鹏谢东阳吴婉银李想
Owner YUNNAN UNIV
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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap