Facial Emotion Recognition Using Deep Learning Method

Status: Finished Date: 2017.11-2017.12 Member: Zexiang Liu, Yushu Li, Yining Heng, Qian Ma

Problem Description:

In this project, we focused on emotion recognition using facial expression with multiclass and multitask CNN methods. Our goal is to develop deep learning Convolutional Neural Network combined with Kernel Support Vector Machine to achieve classifications, and compare its performance with other two methods(CNN and Multi-task CNN Model).

Our solution and the classification results please refer to the full project report.

EECS545 Final Report.pdf