Short video recommendation from locally stored videos, using facial recognition and rule based mapping for classifying into classes. To leverage ML techniques for the purpose of mood-based short video recommendations. Predict the facial emotion of a person using FER model. Use the recommendation filter to suggest appropriate short videos.
Fine-tuned pre-trained VGG16 model for extracting facial emotion features and implemented rule-based mapping for the recommendation of appropriate videos. Dealt with seven emotions facial image data and achieved an accuracy of 86%.
