Challenges of Facial Micro-Expression Detection and Recognition: A Survey

Abstract

Facial expressions are mainly divided into two broad categories micro-expression and macro-expression. During human interaction, both macro-expression and micro-expressions, as well as intermediate expressions, are present. Macro gestures are deliberate in design and cover large areas of the face. The duration of macro-expression is 1/2 to 4 s, while micro expressions are automatic in nature, having a period of 65 milliseconds to 500 milliseconds and revealing the mind’s genuine emotions. Micro-expression covers minimal face area. Detection of macro-expression is pretty much straightforward and easily identified due to the short retention period. A lot of research has been done for macro expression recognition, and almost 95% accuracy has been achieved. A significant part of the research has also been done for micro-expression recognition by the extension of the macro-expression recognition approach, but still, the researcher lacks to achieve better accuracy. This research paper discusses the challenges of micro-expression detection and recognition. Few methods that cover these challenges are also discussed, but these methods also have some limitations that can be considered for future work. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.