Facial Emotion Recognition Using Conventional Machine Learning and Deep Learning Methods: Current Achievements, Analysis and Remaining Challenges

Facial emotion recognition (FER) is an emerging and significant research area in the pattern domain. In daily life, role of non-verbal communication significant, overall communication, its involvement around 55% to 93%. analysis efficiently used surveillance videos, expression analysis, gesture recognition, smart homes, computer games, depression treatment, patient monitoring, anxiety, detecting lies, psychoanalysis, paralinguistic operator fatigue robotics. this paper, we present a detailed review on FER. The literature collected from different reputable published during current decade. This based conventional machine learning (ML) various deep (DL) approaches. Further, FER datasets for evaluation metrics that are publicly available discussed compared with benchmark results. paper provides holistic using traditional ML DL methods highlight future gap domain new researchers. Finally, work guidebook very helpful young researchers area, providing general understating basic knowledge state-of-the-art methods, experienced looking productive directions work.