Open AccessA Real-time Face Tracking and Recognition System Based on SeetaFace
Author(s) -
Dilizhati Yilihamu,
Palidan Tuerxun,
Abdusalam Dawut,
Askar Hamdulla
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1673/1/012043
Subject(s) - computer science , artificial intelligence , computer vision , facial recognition system , face (sociological concept) , facial motion capture , tracking (education) , upload , face detection , tracking system , three dimensional face recognition , function (biology) , object class detection , pattern recognition (psychology) , kalman filter , psychology , social science , pedagogy , evolutionary biology , sociology , biology , operating system
A real-time face tracking and recognition system is constructed based on Seetaface, and it is proposed to use Gamma correction to reduce the impact of illumination changes on the detection and recognition results. First, the system has face detection function, then by using OpenCV to get the image in front of the camera, the system completes the function of real-time face tracking, users can also upload local images to make face comparison. If a stranger’s face is detected, the system can collect information and train it. The system achieves multiple face tracking as well as recognition. Through the analysis and evaluation on public datasets. The improved face tracking and recognition system has a satisfactory accuracy and performance.