Facial Emotion Prediction Using Transfer Learning and Deep Learning Models

Emotion prediction from face is to be monitored frequently in class room, public crowd are and social media. This monitoring can help to correct any suspicious activity on flat forms. Emerging area from the artificial intelligence is deep learning as well as transfer learning to predict various tasks. Simply put, emotion is a person's state of feeling or mood. It can be influenced by a person's situation or mood, and as technology has advanced, there have been creation of computer-based facial recognition systems that can determine a person's emotional state by analyzing their facial expressions in pictures or videos. Proposed research is used to find face emotions using transfer learning and deep learning. Convolutional neural networks (CNNs), MobileNetv2 and Vision Transformer (ViT) play an important role in predicticting facial emotions such as happy, sad, angry, surprise, fear and neutral Proposed method compared transfer learning model with deep learning model and got 94% accuracy which is higher than other methods.

Перевод пока недоступен