How to protect biological privacy in the AI era? Overview of the latest "privacy enhancement technology in biometrics" of the Autonomous University of Madrid, comprehensively detailing the biometric p

Privacy protection is now AI Technology application is a problem that we are concerned about . Recently, scholars from the Autonomous University of Madrid in Spain and the University of Darmstadt in Germany released 《 A survey of privacy enhancement techniques in biometrics 》, It provides a conceptual overview of biometric privacy enhancement technology under a unified framework , Very noteworthy ！

Thesis link ：https://arxiv.org/abs/2206.10465

Privacy enhancement technology is a technology to realize basic data protection principles . In biometrics , Different types of enhanced privacy technologies have been adopted , To protect stored biometric data , These data are usually classified as sensitive data . In this regard , Various taxonomies and conceptual classifications are proposed , And carried out standardization . However , These efforts have focused on certain subcategory technologies that enhance privacy , Therefore, it lacks universality . this This work provides a conceptual overview of biometric privacy enhancement technology under a unified framework . Key aspects and differences between existing concepts will be highlighted in detail in each processing step . The basic properties and limitations of the existing methods are discussed , And related to data protection technology and principles . Besides , The scenario and method of biometric privacy enhancement technology evaluation are also proposed . This paper aims to serve as a starting point in the field of biometric data protection , For experienced researchers and non experts .

Storing unprotected biometric data representations and unprotected biometric templates leads to privacy issues . In the picture , We consider face images . Similar concerns apply to other biometrics .