Facial expression recognition study based on convolutional neural network

Facial expression recognition study based on convolutional neural network

The author of this article intends to study facial expression recognition based on deep neural networks. The first part introduces the traditional methods of facial expression recognition. Then introduces the technical development of deep neural network in the field of image recognition. The second...

Full description

Saved in:
Bibliographic Details
Main Author: Yu, Zhen Lin
Other Authors: Lu Shijian
Format: Final Year Project
Language:English
Published: 2019
Subjects:
DRNTU::Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
Online Access:http://hdl.handle.net/10356/77044
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:The author of this article intends to study facial expression recognition based on deep neural networks. The first part introduces the traditional methods of facial expression recognition. Then introduces the technical development of deep neural network in the field of image recognition. The second part discusses the specific implementation of this project. The facial expression recognition was implemented with Python in Pytorch Framework. The comparative analysis was based on 2 convolutional neural networks (VGG19 and Resnet18) implemented on 2 databases. The compare results show Resnet18 achieved better performance than VGG19. After comparison the article will test visualization examples using Resnet18 model. At last the article will discuss recommendations of future work.

Similar Items