Automatic face labeling method and system

Abstract

The invention discloses an automatic face labeling method and system. The automatic face labeling method includes: step S1, labeling a current face image to obtain a labeling result; step S2, displaying the labeling result, judging whether the labeling result meets requirements, and re-labeling the current face image in a case, where the labeling result does not meet the requirements, until the requirements are met; and step S3, repeatedly executing the step S1 and the step S2 on a next face image in a case, where the current face image meets the requirements, until labeling of all face imagesends. According to the method, faces appearing in videos or pictures can be more accurately labeled through the above technical solution, requirements of a face re-recognition algorithm are satisfied, and better data are provided for upgrading the algorithm.

Description

technical field [0001] The present invention relates to the field of image processing, in particular to a method and system for automatic face labeling. Background technique [0002] With the development of modern computer technology, face recognition technology has been widely used in security verification, human-computer communication, public security system, etc., and also plays a great role in video conferencing, file management, medical treatment, etc. Face recognition technology has gradually become a key topic of research. [0003] While humans can remember and recognize thousands of different faces, and recognize faces and their expressions effortlessly, automatic face recognition by machines is extremely challenging, not least because The facial structure of the human face is complex, and the movement of the facial muscles makes the human face a non-rigid object. Human faces will change with age, and human beings have rich expressions, and they will also be affect...

AI Technical Summary

Problems solved by technology

[0004] To sum up, it is very important to develop a better face recognition algorithm. There are many annotation systems on the market that can mark faces in pictures or videos, but they cannot meet the accuracy of face annotation.

In addition, most of the face annotation tool systems on the market are aimed at technical developers, which are difficult to operate and master

[0005] For the problems in related technologies, no effective solutions have been proposed yet

Images