• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Konferenzschrift
  4. MADation: Face Morphing Attack Detection with Foundation Models
 
  • Details
  • Full
Options
2025
Conference Paper
Title

MADation: Face Morphing Attack Detection with Foundation Models

Abstract
Despite the considerable performance improvements of face recognition algorithms in recent years, the same scientific advances responsible for this progress can also be used to create efficient ways to attack them, posing a threat to their secure deployment. Morphing attack detection (MAD) systems aim to detect a specific type of threat, morphing attacks, at an early stage, preventing them from being considered for verification in critical processes. Foundation models (FM) learn from extensive amounts of unlabelled data, achieving remarkable zero-shot generalization to unseen domains. Although this generalization capacity might be weak when dealing with domain-specific downstream tasks such as MAD, FMs can easily adapt to these settings while retaining the built-in knowledge acquired during pretraining. In this work, we recognize the potential of FMs to perform well in the MAD task when properly adapted to its specificities. To this end, we adapt FM CLIP architectures with LoRA weights while simultaneously training a classification header. The proposed framework, MADation surpasses our alternative FM and transformer-based frameworks and constitutes the first adaption of FMs to the MAD task. MADation presents competitive results with current MAD solutions in the literature and even surpasses them in several evaluation scenarios. To encourage reproducibility and facilitate further research in MAD, we publicly release the implementation of MADation at https://github.com/gurayozgur/MADation.
Author(s)
Loureiro Caldeira, Maria Eduarda
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Ozgur, Guray
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Chettaoui, Tahar
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Ivanovska, Marija
University of Ljubljana  
Peer, Peter
University of Ljubljana  
Boutros, Fadi  orcid-logo
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Struc, Vitomir
University of Ljubljana  
Damer, Naser  
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Mainwork
IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2025. Proceedings  
Project(s)
Next Generation Biometric Systems  
Next Generation Biometric Systems  
Funder
Bundesministerium für Bildung und Forschung -BMBF-  
Hessisches Ministerium für Wissenschaft und Kunst -HMWK-  
Conference
Winter Conference on Applications of Computer Vision 2025  
Workshop on Manipulation, Generative, Adversarial, and Presentation Attack in Biometrics 2025  
Open Access
DOI
10.1109/WACVW65960.2025.00179
Language
English
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Keyword(s)
  • Branche: Information Technology

  • Research Line: Computer vision (CV)

  • Research Line: Human computer interaction (HCI)

  • Research Line: Machine learning (ML)

  • LTA: Interactive decision-making support and assistance systems

  • LTA: Machine intelligence, algorithms, and data structures (incl. semantics)

  • Biometrics

  • Face recognition

  • Machine learning

  • Morphing attack

  • ATHENE

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024