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

Method and system for face recognition

A face image and block technology, applied in the field of face recognition, can solve the problems of inability to capture structural information, sensitivity to noise and outliers, etc.

Active Publication Date: 2020-04-24
BEIJING SENSETIME TECH DEV CO LTD
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Current measurement methods cannot meet all these challenges
Many methods based on similarity (dissimilarity) measurements directly use pairwise distances to calculate the similarity (dissimilarity) between faces, and these methods cannot capture structural information for high-quality discrimination
Although some studies apply said structural information to the measurements, in general the developed algorithms are sensitive to noise and outliers

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
  • Method and system for face recognition
  • Method and system for face recognition
  • Method and system for face recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0078] Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When appropriate, the same reference numerals are used in all drawings to denote the same or similar parts.

[0079] figure 1 Illustrated is a flowchart of a process 100 for recognizing a human face according to an embodiment of the present application, and the process will be discussed in detail below.

[0080] At step S101

[0081] Such as figure 1 As shown in, a pair of face images is obtained from a face image database, and then each of the face images is divided into a plurality of image blocks, where each block in one image and the other image Corresponding blocks in form a pair of blocks. Such as figure 2 As shown in a, face a and face b form a pair of face images, and block p in face a and block q in face b form a pair of blocks.

[0082] At step S102

[0083] In this step, the first similarity of each pair of blocks will be determined throu...

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 system for recognizing human faces. The method may include: acquiring a pair of face images; dividing each of the acquired face images into a plurality of image blocks, wherein each block in one image is associated with a corresponding block in the other image Form a pair of blocks; determine the first similarity of each pair of blocks; determine the second similarity of the pair of human face images from all block pairs; and fuse the first similarity determined for each pair of blocks degree and the second degree of similarity determined for the pair of face images.

Description

Technical field [0001] This application relates to a method and system for identifying human faces. Background technique [0002] Face recognition has been extensively studied in the past two decades. However, large individual internal changes, such as changes in posture, lighting, and expression, are still challenging for robust face recognition in real life photos. [0003] Traditional measurement methods for face recognition have several limitations, which limit their wider application in situations of large individual internal changes. For example, a method for diversified distribution of face images for a person has been disclosed in the art. In fact, due to the high complexity in face data, these distributions usually have different densities, sizes, and shapes. In addition, the face space often contains noise and outliers. Current measurement methods cannot cope with all these challenges. Many methods based on similarity (dissimilarity) measurement directly use pairwise...

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): G06K9/00G06V10/426G06V10/764
CPCG06V40/171G06V10/426G06V10/764G06F18/24147G06F18/22G06V40/162G06V40/172
Inventor 汤晓鸥陆超超赵德丽
Owner BEIJING SENSETIME TECH DEV CO LTD