A Study on the Emoticon Extraction based on Facial Expression Recognition using Artificial Intelligence

• 발행기관: 한국유통과학회
• 저자명: Bong Jae Jeong Min Soo Kang
• 간행물 정보: 『한국유통과학회 학술대회 논문집』2017년 동계 국제학술대회, 237~239쪽, 전체 3쪽
• 주제분류: 경제경영 > 경제학
• 파일형태: PDF
• 발행일자: 2017.12.27

영문 초록

In this paper, I proposed a prototype model that extracts the same emoticon after facial expression recognition in Android smart device. Understanding and expressing facial expressions are important for the interaction between people and computers. Technologies for recognizing human facial expressions are gaining popularity. It is a useful technology because you can use the emoticons you want only with face recognition by using the camera without searching the emoticons one by one used by users. The technology proposed in this paper uses ‘FER2013’ dataset offered by Kaggle website and based on Tensorflow library for improving facial expression recognition accuracy CNN(Convolutional Neural Network) algorithm is used to make a recognition model. Using this model, facial expression is recognized by Android Smart Device and extracts the same emoticon as a result.

목차

Abstract

1. Introduction

2. Related Studies

3. Body

4. Results

5. Conclusion

References

키워드

Deep Learning Tensorflow CNN(Convolutional Neural Network) Algorithm Facial Expression Recognition Anrdoid Smart Device