Design of robust security systems has gained significant progress in the past decade, motivated by the need to fight crime and terrorism, to strengthen security through border management, and provide cyber security. Face and gesture analysis have played crucial roles in related video-based surveillance and authentication, complementing traditional methods. In spite of the recent advancements, including (a) the incorporation of face analysis in large-scale identification programs, (b) skyrocketing performance rates of deep learning based face analysis algorithms, as well as (c) the deployment on personal mobile devices, there are a number of remaining open challenges to address, as reported recently [1]. Such open challenges include fundamental questions w.r.t. distinctiveness and persistence of biometric traits, design of data-driven methods streamlined for mobile devices, presentation attack detection, heterogeneous matching systems, soft biometrics, privacy, biases in biometrics, as well as behavioural biometrics.

We plan to host this special session with a wider scope to push forward and document recent advancements in addressing the above challenges. The goal of this special session is to bring together researchers and practitioners working in the area of biometrics, security and privacy, and to address a wide range of theoretical and practical issues related to systems in the context of security. This session will aim to open a debate on new opportunities and new challenges in the area, and at unifying the efforts toward the development of new adequate tools, protocols and databases for evaluating and monitoring the progress in the field.

Topics of interest include, but are not limited to:

  • Video based surveillance using face and gesture analysis
  • Design and testing of multi-factor authentication systems
  • Deep learning based presentation attack detection
  • Generalization of presentation attack detection methods to real-world scenarios and systems
  • Detecting DeepFakes
  • Person Re-identification in the Wild
  • Behaviour monitoring
  • Face analysis for lie or deceit detection
  • Mal-intent detection using behaviour patterns
  • Databases on biometric technology, lie detection, and behaviour pattern analysis
  • Biometric technology for mobile devices at distance
  • Information fusion of biometric modalities
  • Privacy and ethics
  • Deployment of biometrics in realistic security scenarios
  • Ubiquitous biometrics
  • Fundamental limits of biometrics (e.g., distinctiveness, permanence)
  • Soft biometrics
  • Open set recognition
  • Fusion and multimodal analysis
  • Usability
  • Heterogeneous matching systems,
  • Biases in biometrics
  • Behavioural biometrics.

Organizers:

  • Antitza Dantcheva, INRIA, Team STARS, Sophia Antipolis, France
  • S L Happy, INRIA, Team STARS, Sophia Antipolis, France
  • Tamara Rezk, INRIA, Team Indes, Sophia Antipolis, France
  • Cunjian Chen, Michigan State University, i-PRoBe Lab, East Lansing, USA