Skip to content
No description, website, or topics provided.
Python Limbo
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Emotion Recognition using Convolutional Neural Network

Repository details:

All the scripts in this repository require the installation of OpenCV, Caffe and Cuda. Please set the respective paths in the script according to your installation.

  1. - train the SVM classifier layer on the features obtained from GoogLeNet. Dataset used in training the model is available upon request.

  2. "models-sigmoid-cv" directory consists of a trained Emotion Classifier mdoel and the Normalizer transform value used on the training set.

  3. - perform Emotion Classification Task on any given input image.

  4. execution command - python "path to your image"

  5. will search for faces in the input image using OpenCV Haar cascade classifier and feeds the face patch into the CNN. The predicted emotion of the face will be written on the output image with bounding box annotation. Output image path - 'predictions/' directory which will be created on runtime.

You can’t perform that action at this time.