Monozygotic twins, commonly known as identical twins, share extreme facial resemblance that they deceive the majority of the prevailing face recognition systems. Hence, the facial recognition systems, which are turning out to be exploited in almost all real-life biometric authentication-needing applications, require significant upgradation to cope with this great challenge. This paper pays special attention towards categorizing and investigating the existing research on the facial recognition of monozygotic twins in an exhaustive sense. Subsequently, the research gap is analyzed. It is then followed by the description of the possible refinements, like the choice of feature and other methodological advancements, which may aid in augmenting the twin recognition accuracy of the face recognition systems. It is believed that the outcomes of the review can well-support and pave the way for further research in the future. Consequently, the "Double Trouble" posed on the commercial face recognition systems can be effectively evaded. (c) 2021 Elsevier B.V. All rights reserved.
展开▼
