A Systematic Review of Facial Expression Recognition (FER)

Abstract

In order to improve human personification among humans, robots, and sophisticated symbols, by utilizing advances in artificial intelligence, Facial Expression Recognition (FER) seeks to identify facial expressions from still images. New problems and methods that are not given much thought in that FER mindset are confronted as the FER section moves from controlled laboratory conditions to increasingly complicated real-world scenarios. Advanced techniques have been developed quickly. This paper offers a thorough analysis of image-based static FER (SFER) methods, looking at everything from challenge-centred order to model-arranged improvement. We start with a basic analysis of current surveys, an overview of common datasets and methodologies, and a workflow using FER in order to build a solid basis.