Implementation of an Emotion Recognition System Based on OpenFace

• Main Authors: CHEN, JIA-HUA, 陳佳華
• Other Authors: HSU, JIA-LIEN
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/bzsnk6

碩士 === 輔仁大學 === 資訊工程學系碩士班 === 106 === The face recognition system has become more and more popular nowadays. It has not only been applied to some of access control systems, but also widely used in our daily life. For example, it can precisely identify facial features in order to unlock a mobile phone. And, it will automatically tag names in photos which have been uploaded to the social media. OpenFace is an open source of the face recognition system, which can instantly perform a facial recognition by using pre-trained models within its database. This dissertation is based on the statistics of quantified face landmarks processed by OpenFace, along with the training of WEKA emotion recognition modules. In addition to analyzing facial patterns toward expression detected, the application of clinical usage can be put in practice.