Kolf, Jan Niklas ; Damer, Naser ; Boutros, Fadi (2024)
GraFIQs: Face Image Quality Assessment Using Gradient Magnitudes.
2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW 2024). Seattle, USA (16.6.2024 - 22.6.2024)
doi: 10.1109/CVPRW63382.2024.00156
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
Face Image Quality Assessment (FIQA) estimates the utility of face images for automated face recognition (FR) systems. We propose in this work a novel approach to assess the quality of face images based on inspecting the required changes in the pre-trained FR model weights to minimize differences between testing samples and the distribution of the FR training dataset. To achieve that, we propose quantifying the discrepancy in Batch Normalization statistics (BNS), including mean and variance, between those recorded during FR training and those obtained by processing testing samples through the pretrained FR model. We then generate gradient magnitudes of pretrained FR weights by backpropagating the BNS through the pretrained model. The cumulative absolute sum of these gradient magnitudes serves as the FIQ for our approach. Through comprehensive experimentation, we demonstrate the effectiveness of our training-free and quality labeling-free approach, achieving competitive performance to recent state-of-the-art FIQA approaches without relying on quality labeling, the need to train regression networks, specialized architectures, or designing and optimizing specific loss functions.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2024 |
| Autor(en): | Kolf, Jan Niklas ; Damer, Naser ; Boutros, Fadi |
| Art des Eintrags: | Bibliographie |
| Titel: | GraFIQs: Face Image Quality Assessment Using Gradient Magnitudes |
| Sprache: | Englisch |
| Publikationsjahr: | 17 September 2024 |
| Verlag: | IEEE |
| Buchtitel: | Proceedings: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops: CVPRW 2024 |
| Veranstaltungstitel: | 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW 2024) |
| Veranstaltungsort: | Seattle, USA |
| Veranstaltungsdatum: | 16.6.2024 - 22.6.2024 |
| DOI: | 10.1109/CVPRW63382.2024.00156 |
| Kurzbeschreibung (Abstract): | Face Image Quality Assessment (FIQA) estimates the utility of face images for automated face recognition (FR) systems. We propose in this work a novel approach to assess the quality of face images based on inspecting the required changes in the pre-trained FR model weights to minimize differences between testing samples and the distribution of the FR training dataset. To achieve that, we propose quantifying the discrepancy in Batch Normalization statistics (BNS), including mean and variance, between those recorded during FR training and those obtained by processing testing samples through the pretrained FR model. We then generate gradient magnitudes of pretrained FR weights by backpropagating the BNS through the pretrained model. The cumulative absolute sum of these gradient magnitudes serves as the FIQ for our approach. Through comprehensive experimentation, we demonstrate the effectiveness of our training-free and quality labeling-free approach, achieving competitive performance to recent state-of-the-art FIQA approaches without relying on quality labeling, the need to train regression networks, specialized architectures, or designing and optimizing specific loss functions. |
| Freie Schlagworte: | Biometrics, Machine learning, Face recognition |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Graphisch-Interaktive Systeme |
| Hinterlegungsdatum: | 08 Okt 2024 09:17 |
| Letzte Änderung: | 08 Okt 2024 09:17 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum