A Review of Automated Micro-expression Analysis

Micro-expression is a type of facial expression that manifested for very short duration. It difficult to recognize the manually because it involves subtle movements. Such expressions often occur unconsciously, and therefore are defined as basis help identify real human emotions. Hence, an automated approach micro-expression recognition has become popular research topic interest recently. Historically, early researches on have utilized traditional machine learning methods, while more recent development focused deep approach. Compared learning, which relies manual feature processing requires use formulated rules, networks produce accurate performances through end-to-end methodology, whereby features were extracted optimally training process, utilizing large set data. This paper reviews developments trends in from earlier studies (hand-crafted approach) present (deep approach). Some important topics will be covered include detection videos, apex frame spotting, well performance discussion reviewed methods. Furthermore, major limitations hamper systems also analyzed, followed by recommendations possible future directions.