jinsung4069 / emotion_classification

Real-Time Facial Expression Recognition Using Webcam (detection + classification)

Emotion Recognition

reference github : https://github.com/XiaoYee/emotion_classification

the purpose of this repository is to explore Degree Of Understanding(DoU) with deep learning in real time

Environment:

    python 3.7.4
    pytorch 1.2.0

Detection Models

1. Haar_cascade
2. Blazeface
3. DNN
4. HOG
5. MMOD
6. MTCNN

Classification Models

1. VGG
2. VGG_GRU
3. RESNET_GRU

Models Detail:

...

Dataset:

---

reclassify FER2013, JAFFE, KDEF

DATASET
│  
└───Not Understand(4,327)
│   
└───Neutral (2,176)
│   
└───Understand(3,378)

Performance:

all the performance train and test on FER2013, JAFFE, KDEF train and validation partition. speed [sec/4frame]

Model	VGG	VGG+GRU	Resnet+GRU
precision	49.8%	82.7%	94.3%

test with private video

Model	Detection speed	Classification speed	sum
blazeface	0.089	0.281	0.370
dnn(opencv)	0.209	0.278	0.487
mmod(dlib)	13.181	0.316	13.50
haar(opencv)	0.244	0.265	0.509
mtcnn	0.648	0.283	0.931
hog(dlib)	0.149	0.301	0.450