Research Area: Machine Learning
The birth of face recognition technology began in the 1960s, and it has experienced a general development process: based on face structure features (1970-1990), statistical features (1991-2000), big data and complex algorithms (2001-present). Among them, the first phase of face recognition technology is mainly to establish a grayscale image model through studying facial features, and at the same time, it can not complete the automatic recognition. In the second stage, multi-dimensional feature vectors are adopted to represent facial features, and previous empirical knowledge should be used for judgment. With the development of artificial intelligence, modern face recognition technology integrates artificial intelligence, machine learning, image processing and other technologies to study the face recognition problems facing with real conditions.This paper will summarize, analyze and discuss some traditional face recognition algorithms and current typical facial recognition technology based on artificial intelligence algorithm, and through the study on facial model building, facial feature representation, algorithm accuracy and other influencing factors, the advantages and disadvantages of each face recognition algorithm when applied to various fields will be analyzed, and then these algorithms will be evaluated and prospected.
Keywords:
Author(s) Name: Enjie Jiang
Journal name:
Conferrence name: International Conference on Computer Vision, Image and Deep Learning (CVIDL)
Publisher name: IEEE
DOI: 10.1109/CVIDL51233.2020.00010
Volume Information:
