A Pytorch Implementation of FER(Facial-Expression-Recognition)
Introduction
This project aims to classify facial expression. Here I provide seven types of expression, including Angry Disgusted Fearful Happy Sad Surprised Neutral. With 250 epochs, this accuracy of baseline achieves 70.382%
Here is the trained model link: ———— Trained Model Link password:5nfw
- Backbone ——VGG16
- Dataset ——FER2013:
Dataset Link 240×240 Data（Train、Val、Test） password:5j3x
Highlight
- In this project, face detection part is applied, which can definitely improve the test accuracy. More over, it can support the robust of the model, especially no face input image.
- GPU and CPU all support. If no GPU, it's OK.
- Dependencies fewer.
- When testing, batch images input is supported in the demo.
Results Show
Requirement
Recommend to use Anaconda
- Ubuntu16.04 (Windows also avaliable)
- Python 3.6
- Pytorch (latest version or old version are all fine)
- torchvision
- numpy
- matplotlib
- opencv(cv2)
- pillow
Dataset Process
FER2013 includes 35887 pictures: 48 × 48 pixels, here using bilinear interpolation to resize the expression pictures to 240 × 240 pixels. The input of the net is 224 × 224, same as original VGG16.
Train
First, put the processed dataset in the folder "data", the data folder like following:
-- data
------- train
------------------ 0
---------------------------00000.jpg
---------------------------00005.jpg
...
------------------ 1
---------------------------00023.jpg
...
...
------------------ 6
---------------------------00061.jpg
...
------- val
------------------ 0
---------------------------00006.jpg
...
------------------ 1
---------------------------00043.jpg
...
...
------------------ 6
---------------------------00021.jpg
...
------- test
------------------ 0
---------------------------00008.jpg
...
------------------ 1
---------------------------00011.jpg
...
...
------------------ 6
---------------------------00022.jpg
...
0-6 represent 7 different expression: Angry Disgusted Fearful Happy Sad Surprised Neutral
Demo
Image Input
python demo_image.py
Running the demo, first need to type the image name, such as 1.jpg.
Put input images in input folder
Camera Detection
python demo_camera.py
Batch Image Input
python demo_image_batch.py
TODO
Find image process methods to improve the accuracy.
Apply RPN face detection to improve accuracy.
Issue and Suggestion
Any questions, open a new issue.