Devinder Sarai devindersarai
devindersarai
/ live-facial-expression-recognition.py
Last active
April 12, 2021 02:20
Facial expression from a webcam (or video file) live.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import cv2 | |
| import numpy as np | |
| import os | |
| import keras | |
| from keras.models import load_model | |
| dicts = ['empty', 'neutral', 'calm', 'happy', 'sad', 'angry', 'fear', 'disgust', 'surprise', 'neutral', 'calm', 'happy', 'sad', 'angry', 'fear', 'disgust', 'surprise'] | |
| # To capture video from a webcam |
devindersarai
/ cnn-model.py
Last active
April 12, 2021 01:42
CNN model for facial expression recognition, an arbitrary dropout value of 0.5 was chosen.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| model = Sequential() | |
| model.add(Convolution2D(filters=16, kernel_size=(7, 7), padding='same', input_shape=(256,256,3))) | |
| model.add(BatchNormalization()) | |
| model.add(Convolution2D(filters=16, kernel_size=(7, 7), padding='same')) | |
| model.add(BatchNormalization()) | |
| model.add(Activation('relu')) | |
| model.add(AveragePooling2D(pool_size=(2, 2), padding='same')) | |
| model.add(Dropout(0.5)) | |
| model.add(Convolution2D(filters=32, kernel_size=(5, 5), padding='same')) |
devindersarai
/ video-preprocessing.py
Created
April 11, 2021 02:34
Preprocessing the videos from the RAVDESS dataset into individual frames of the actors faces and saving them to the appropriate folder (based on the labeled emotion).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| dim = (384, 216) | |
| dim_final = (224, 224) | |
| frameRate = 30 # Every nth frame is saved. | |
| # Detect the face in image using HAAR cascade, crop it, resize it, and finally save it. | |
| for actor in glob.glob(VIDEO_FOLDER): | |
| for videoFile in glob.glob(actor + '/*'): | |
| print(videoFile) | |
| emotion = label.get_emotion(videoFile[-24:]) | |
| cap = cv2.VideoCapture(videoFile) # capturing the video from the given path | |
| while(cap.isOpened()): |