Recognizing facial expressions from images, videos and real-time stream
Cerebro is a python package for Facial Expression Detection, we provide a trained model with accuracy around 98% of 8 emotions [Happy, surprise, contempt, sad ,angry,disgust,Neutral,Fear], with a very simple interface for detection from image,video with any rotation and Real time streaming.
Example In this Example we get an image , predict an emotion then save it with the emotion.
from interface import video_stream as vs from interface import process_image as pi def main(): im = cv2.imread("interface/7.jpg") items =pi.extract_faces_emotions(im) im =pi.mark_faces_emotions(im) cv2.imwrite("interface/77.jpg",im) cv2.imshow("detected emotions",im) cv2.waitKey(0) if __name__ == '__main__': main()
Installition Cerebro depends on some python packages, once you install Cerebro any missing Module will be automatically installed, for FFmpeg use this link.
Installation by hand: download the sources, either from PyPI or, if you want the development version, from GitHub, clone the project then use this command in terminal to setup.
$ (sudo) python setup.py install
Installation with pip: if you have pip installed, just type this in a terminal:
$ (sudo) pip install CEREBRO8
Using Model : once You install Cerebro You have to dowenload our trained model from this link and full model link then add them to new foldercalled
Using Landmark : if you want to use Landmark feature extractor you have to dowenload this file landmarks with 68 point using this link in this path
saved-models/face-landmarks "create new folder called landmarks in saved-models"
Real Time demo
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size CEREBRO8-0.0.1-py3-none-any.whl (31.8 kB)||File type Wheel||Python version py3||Upload date||Hashes View hashes|
|Filename, size CEREBRO8-0.0.1.tar.gz (14.0 kB)||File type Source||Python version None||Upload date||Hashes View hashes|