• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Anderes
  4. PreFIQs: Face Image Quality Is What Survives Pruning
 
  • Details
  • Full
Options
2026
Paper (Preprint, Research Paper, Review Paper, White Paper, etc.)
Title

PreFIQs: Face Image Quality Is What Survives Pruning

Title Supplement
Published on arXiv
Abstract
Face Image Quality Assessment (FIQA) evaluates the utility of a face image for automated face recognition (FR) systems. In this work, we propose PreFIQs, an unsupervised and training-free FIQA framework grounded in the Pruning Identified Exemplar (PIE) hypothesis. We hypothesize that low-utility face images rely disproportionately on fragile network parameters, resulting in larger geometric displacement of their embeddings under model sparsification. Accordingly, PreFIQs quantifies image utility as the Euclidean distance between L2-normalized embeddings extracted from a pre-trained FR model and its pruned counterpart. We provide a first-order theoretical justification via a Jacobian-vector product analysis, demonstrating that this empirical drift serves as a computationally efficient approximation of the exact geometric sensitivity of the latent embedding manifold. Extensive experiments across eight benchmarks and four FR models demonstrate that PreFIQs achieves competitive or superior performance compared to state-of-the-art FIQA methods, including establishing new state-of-the-art results on several benchmarks, without any training or supervision. These results validate parameter sparsification as a principled and practically efficient signal for face image utility, and demonstrate that quality is, in essence, what survives pruning.
Author(s)
Kolf, Jan Niklas  
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Ozgur, Guray
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Atzori, Andrea
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Babnik, Žiga
Univ. Ljubljana
Štruc, Vitomir
Univ. Ljubljana
Damer, Naser  
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Boutros, Fadi  orcid-logo
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Conference
Conference on Computer Vision and Pattern Recognition Workshops 2026  
Open Access
File(s)
Download (3.39 MB)
Rights
CC BY-NC-SA 4.0: Creative Commons Attribution-NonCommercial-ShareAlike
DOI
10.48550/arXiv.2605.13396
10.24406/publica-9876
Language
English
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Keyword(s)
  • Face Image Quality Assessment

  • Face Recognition

  • Model Pruning

  • Branche: Infrastructure and Public Services

  • Research Line: Computer vision (CV)

  • Research Line: Machine learning (ML)

  • ATHENE

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