Emotional Response
Published:
A deep learning model that analyses photos to determine the emotional response of a human seeing the photo.
The model uses images of artwork, trained on 8 basic human emotions: anger, fear, sadness, disgust, surprise, anxiety, trust, and joy. Images that evoke these feelings were used to train a Convolutional Neural Network to predict the emotions humans would feel on viewing a piece of art or a photograph.
The results were very low across all 8 classes, so the classifications were reduced to binary (happy vs. sad), which achieved significantly better results.