Skip to content

Latest commit

 

History

History
63 lines (37 loc) · 2 KB

File metadata and controls

63 lines (37 loc) · 2 KB

Facial_Expression_Recognition_FER2013

Powered by Python 3.5

Library Requirements

 pip install tensorflow-gpu
 pip install keras
 pip install numpy
 pip install sklearn
 pip install pandas
 pip install opencv-python==3.3.1

Dataset

FER2013

More details : Kaggle Challenge - https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge/data

Accuracy Achivement

Kaggle Challenge Winner 71.2%.

Model Combine CNN1 + CNN2 + CNN_FAST + CNN_SIFT 72.89%

My Model

Confusion Matrix

Confusion Matrix on Private Test Set FER2013

CNN Architectures

Build from scratch in /cnn folder .

HOW ARE YOU TODAY? Application

Run UI.py file to open application.

User Interface

You can choose some image and video that include human faces to detect emotions

Click Button "Open Your Camera" for real-time detecting yourself emotions from webcam.

Highlight Results

thor

steve

lookingbackmeme

soilder

ron

Video Realtime

Examples will be uploaded later.

References

The Main Idea From This Article.