Abstract
|
Emotion recognition has become an area of research interest in recent years because of its applicability in several domains like neuropathy treatment, online shopping, mental rehabilitation, therapeutic gaming, and drug testing, etc. Many researchers proposed techniques for emotion recognition. This review paper focuses on three critical points in emotion recognition using EEG and facial expressions data. First, the mechanisms for capturing emotions using EEG signals have been highlighted. Secondly, an overview of existing techniques either based on handcrafted features or deep learning has been presented. Thirdly, the description and analysis of the databases that are available to validate the performance of algorithms, de-signed for recognizing certain types of emotions have been presented. We have provided an overview and analysis of the research work on emotion recognition from 2017 to 2022. In recent years, there has been a shift from handcrafted features to deep learning. The techniques based on deep learning and handcrafted features have been compared and their strengths and limitations have been elaborated. Finally, future research directions have been highlighted.
|