Skip to content

A lightweight system that predicts and displays emojis from facial emotions in a live webcam feed.

Notifications You must be signed in to change notification settings

004Ajay/Emotion2Emoji

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Emotion2Emoji

A lightweight system that displays emojis corresponding to the facial emotion in a live webcam feed.


Run this application on your system:

Download this repository:

git clone https://github.com/004Ajay/Emotion2Emoji.git

Move to the project folder:

cd Emotion2Emoji

Install required libraries:

pip install tensorflow, deepface, numpy, opencv-python

Initial run may take some time as the above libraries need to download models and weights related to recognition.

Run the code:

python main.py

System Design:


Features:

  • Real-time Responsiveness: The system works spontaneously without a setback, providing instant feedback for a seamless user experience.

  • Multi-face Detection: No face goes unnoticed! The project can detect multiple faces simultaneously, catering to diverse use cases.

  • Stability: The system is designed for stability, ensuring a reliable performance every time.


Drawbacks:

  • When head is tilted sideways the face detection won't work, I think this is a problem with Haarcascade Classifier.

  • Even though the system is very stable most of the time, the system randomly fails for unknown reasons.

  • The system works perfectly when there is a single face in detection, but fails to give the correct emotions when there are multiple faces. Anyway the system show equal number of emojis as number of faces in detection.


See demo video here

About

A lightweight system that predicts and displays emojis from facial emotions in a live webcam feed.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages