Multimodal decomposition with magnification on micro-expressions and its impact on facial biometric recognition

With the advancement of computer vision techniques, genuine human emotions may be captured in split second of facial muscle movements despite of cheating and concealment. These movements are called the micro-expressions. In recent, the study of micro-expressions had been extensively done as research have shown its usefulness in lie detection and clinical diagnosis. Due to its low intensity, sparse and short duration nature, boosting and magnification have applied aiming to improve detection and recognition of micro-expressions. In this paper, we present method to magnify intensity of micro-expressions which inclusive study of its impact on facial biometric recognition.