Differential privacy human body anonymous synthesis method with consistent availability

A technology of differential privacy and synthesis method, which is applied in the field of anonymous synthesis of differential privacy human body, which can solve the problems of human body anonymity and achieve the effect of avoiding collection and abuse

Pending Publication Date: 2022-04-12
HANGZHOU DIANZI UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Most of the current research is based on face anonymity, but this is not enough for human body anonymity, because the human body includes a lot of identity information in addition to the face

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
  • Differential privacy human body anonymous synthesis method with consistent availability
  • Differential privacy human body anonymous synthesis method with consistent availability
  • Differential privacy human body anonymous synthesis method with consistent availability

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] The present invention will be further described below in conjunction with accompanying drawing.

[0059] For a given source image First use step 2 to assign the delegation image corresponding to the delegation identity Then replace the I with the UPT network x The human body in , get the anonymous target image

[0060] The process flow of the differential privacy human body anonymous synthesis method with consistent availability is as follows: figure 1 shown. For a given input human body image First assign a delegated identity (DPA) to obtain a delegated body image Random identity mapping via differential privacy. Then perform Usability Preserving Pose Transformation Network (UPT) to replace I x The human body in , get the anonymous target image The DPA process consists of two steps: building attribute consistent candidate sets (ACCS) and selecting differentially private delegated bodies (DPDP). UPT Network to P x 、A x , B x , M x , s d , The inpu...

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 differential privacy human body anonymous synthesis method with consistent availability. The method is characterized in that an anonymous human body data set is obtained from a source human body data set through a delegated human body data set and an availability keeping posture conversion network; the method comprises the following specific steps: 1, preprocessing; step 2, assigning a delegation identity; step 3, constructing an availability keeping posture conversion network; step 4, training constraint; and 5, model training and testing. According to the method, a new anonymous human body image can be generated, privacy data is prevented from being collected and abused, and the anonymous image can keep the availability of original data. In addition, natural images can be generated under different privacy levels, and human body privacy and data availability are kept.

Description

technical field [0001] The invention relates to the field of image privacy protection, and proposes a differentially private human body anonymous synthesis method with consistent usability. Background technique [0002] The unprecedented application of image capture devices (such as smartphones and surveillance cameras) has resulted in a large number of private human body images in private as well as public areas and these images are easily captured if they are released on the Internet. At the same time, advanced artificial technologies (such as behavior analysis, pedestrian re-identification, face recognition, etc.) can use these data to produce applications with superior performance and play an important role in today's society. However, these applications will be abused by criminals and cause serious privacy concerns. Leakage problem. [0003] To address the privacy concerns of the appeal, some countries have enacted increasingly strict laws and even delisted public data...

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): G06V40/10G06V10/26G06V10/40G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
Inventor 匡振中滕龙斌李忠金俞俊
Owner HANGZHOU DIANZI 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