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

Deep convolutional network for human face identification in complex scene and learning method

A complex scene, deep convolution technology, applied in the computer field, can solve the problems of uneven level of photographers and failure to meet application requirements

Inactive Publication Date: 2017-09-22
上海荷福人工智能科技(集团)有限公司
View PDF3 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, with the rapid increase of digital images including human faces taken from various environments, and the level of the photographers is uneven, the early face recognition technology as mentioned above can be used in such real environments. The recognition accuracy drops rapidly, which is far from meeting the actual application requirements

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
  • Deep convolutional network for human face identification in complex scene and learning method
  • Deep convolutional network for human face identification in complex scene and learning method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0056] A deep convolutional network for face recognition in complex scenes, such as figure 1 As shown, it includes multiple convolution modules and fully connected modules. Each convolution module or fully connected module is composed of convolution kernels, activation functions and pooling functions at different scales. The first three convolution modules are respectively It consists of a convolutional layer, a relu layer and a max pooling layer. The max pool...

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 deep convolutional network for human face identification in a complex scene and a learning method. The network structurally consists of four full convolutional modules and three full connection modules; and parameters of the deep convolutional network are optimized by utilizing a stochastic gradient descent algorithm. The designed deep convolutional network for human face identification is expected to have stable identification performance under multiple view angles for complex illumination in a real application scene; and engineering application demands are met based on relatively high calculation degree under the support of a high-performance GPU.

Description

technical field [0001] The invention relates to a deep convolutional network and a learning method for face recognition in complex scenes, belonging to the field of computer technology. Background technique [0002] The early face recognition algorithms mainly include image preprocessing, proposing traditional face features such as LBP, and applying vector machines to learn and classify the extracted features. Among them, image preprocessing algorithms are particularly critical, including preprocessing algorithms for different lighting conditions, denoising algorithms for image noise, and alignment algorithms for face objects from different angles. After being optimized for a certain face recognition database, such as the classic LFW face recognition database, it has achieved results similar to human recognition levels. [0003] In recent years, with the rapid increase of a wide variety of shooting equipment, a large number of exponentially growing face images appear in var...

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): G06K9/00G06K9/40G06K9/66G06T5/00G06T5/10
CPCG06T5/10G06T2207/20016G06V40/168G06V10/30G06V30/194G06T5/70
Inventor 唐良智王兵魏湘臣
Owner 上海荷福人工智能科技(集团)有限公司
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