TReSPAsS

TRaining in Secure and PrivAcy-preserving biometricS (TReSPAsS)

To combat rising security challenges, the global market for biometric technologies is growing at a fast pace. It includes all processes used to recognise, authenticate and identify persons based on physical and/or behavioural characteristics. The EU-funded TReSPAsS-ETN project will deliver a new type of security protection (through generalised presentation attack detection (PAD) technologies) and privacy preservation (through computationally feasible encryption solutions). The TReSPAsS-ETN Marie Skłodowska-Curie early training network will couple specific technical and transferable skills training including entrepreneurship, innovation, creativity, management and communications with secondments to industry. A cohort of early stage researchers will be equipped with the necessary skill set and capacity to innovate the next generation of secure and privacy-preserving biometrics technologies.

In the course of the TRaining in Secure and PrivAcy-preserving biometricS (TReSPAsS) Marie Skłodowska-Curie Action the Biometrics Research Group of the University of Applied Sciences Darmstadt is currently hosting two PhD candidates:

    1. Privacy preserving indexing of large-scale biometric databases
      Project Description: Large-scale biometric deployments are quickly becoming ubiquitous even though a centralised mass storage of biometric data causes legal issues and privacy concerns. The computational workload of the conventional retrieval method, which requires an exhaustive search in the identification mode, quickly becomes impractical for large systems. This issue necessitates the research into algorithms for efficient biometric identification, i.e. biometric indexing. The feasibility of employing biometric indexing techniques in conjunction with privacy-preserving template protection schemes has yet to be explored. The focus of this project is to fuse both technologies for face, fingerprint, iris, and multi-biometrics in order to achieve computationally efficient and privacy-preserving authentication in large-scale biometric databases.
      Expected Results: Benchmark on the applicability of template protection methods in biometric identification systems; proposal of new template protection techniques which allow for rapid comparison; new indexing techniques designed to index and retrieve protected biometric templates.
    2. Detection of manipulated face images using machine learning methods and image forensics
      Project Description: Recent research has demonstrated the vulnerability of face recognition systems to attacks based on manipulated biometric images. If manipulated biometric images, e.g. morphed images or deep-fakes, are infiltrated to a biometric recognition system accurate and reliable recognition can not be guaranteed. The aim of this project is to develop reliable methods to detect manipulated face images which remains an unsolved research challenge. Deep learning methods will be employed to train appropriate classifiers and the potential of forensic image analysis will be investigated to reliably detected anomalies in facial images.
      Expected Results: Detailed study on the vulnerability of commercial face recognition systems to attacks based on manipulated images; novel techniques to detect manipulated face images; increased understanding of the inter-relation between biometric variance, entropy and the threat of attacks based on manipulated face images.