Face Recognition Based on Facial Landmarks

2017

Closed access Metadata (CSL-JSON)
DOI 10.1109/icaict.2017.8687015
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

These days exists a lot of different approaches for face recognition. In this paper presented one of the traditional approaches with steps which are basis for face recognition. Method used for face recognition includes four steps, which are face detection, head position estimation, features extraction and testing classifier. Automatic detection of face and facial landmarks is very crucial in this method because of the ratio distances between these landmarks used as features for classifier. All the paper long, one by one presented each of four steps. In experiments on a small face database, evaluated accuracy of the method.