arXiv:2505.03610v2 Announce Type: replace 
Abstract: 3D mask presentation attack detection is crucial for protecting face recognition systems against the rising threat of 3D mask attacks. While most existing methods utilize multimodal features or remote photoplethysmography (rPPG) signals to distinguish between real faces and 3D masks, they face significant challenges, such as the high costs associated with multimodal sensors and limited generalization ability. Detection-related text descriptions offer concise, universal information and are cost-effective to obtain. However, the potential of vision-language multimodal features for 3D mask presentation attack detection remains unexplored. In this paper, we propose a novel knowledge-based prompt learning framework to explore the strong generalization capability of vision-language models for 3D mask presentation attack detection. Specifically, our approach incorporates entities and triples from knowledge graphs into the prompt learning process, generating fine-grained, task-specific explicit prompts that effectively harness the knowledge embedded in pre-trained vision-language models. Furthermore, considering different input images may emphasize distinct knowledge graph elements, we introduce a visual-specific knowledge filter based on an attention mechanism to refine relevant elements according to the visual context. Additionally, we leverage causal graph theory insights into the prompt learning process to further enhance the generalization ability of our method. During training, a spurious correlation elimination paradigm is employed, which removes category-irrelevant local image patches using guidance from knowledge-based text features, fostering the learning of generalized causal prompts that align with category-relevant local patches. Experimental results demonstrate that the proposed method achieves state-of-the-art intra- and cross-scenario detection performance on benchmark datasets.

تسلط دراسة حديثة الضوء على أهمية تطوير طرق قوية للكشف عن هجمات تقديم الأقنعة ثلاثية الأبعاد، التي تشكل تهديدًا كبيرًا لأنظمة التعرف على الوجه. من خلال التركيز على المطالبات المعتمدة على المعرفة، يهدف الباحثون إلى تحسين دقة الكشف مع معالجة التحديات المتعلقة بالتكاليف العالية والقدرة المحدودة على التعميم التي تواجهها الأساليب الحالية. هذه الخطوة مهمة لأنها تعزز تدابير الأمان في تطبيقات متعددة، مما يضمن بقاء تقنية التعرف على الوجه موثوقة وفعالة.

Un estudio reciente destaca la importancia de desarrollar métodos robustos para detectar ataques de presentación de máscaras 3D, que representan una amenaza significativa para los sistemas de reconocimiento facial. Al centrarse en indicaciones basadas en el conocimiento, los investigadores buscan mejorar la precisión de la detección mientras abordan los desafíos de los altos costos y la capacidad de generalización limitada que enfrentan los métodos existentes. Este avance es crucial, ya que refuerza las medidas de seguridad en diversas aplicaciones, asegurando que la tecnología de reconocimiento facial siga siendo confiable y efectiva.

Une étude récente souligne l'importance de développer des méthodes robustes pour détecter les attaques par présentation de masques 3D, qui représentent une menace significative pour les systèmes de reconnaissance faciale. En se concentrant sur des invites basées sur les connaissances, les chercheurs visent à améliorer la précision de la détection tout en abordant les défis des coûts élevés et de la généralisation limitée auxquels sont confrontées les méthodes existantes. Cette avancée est cruciale car elle renforce les mesures de sécurité dans diverses applications, garantissant que la technologie de reconnaissance faciale reste fiable et efficace.

A recent study highlights the importance of developing robust methods for detecting 3D mask presentation attacks, which pose a significant threat to face recognition systems. By focusing on knowledge-based prompts, researchers aim to improve detection accuracy while addressing the challenges of high costs and limited generalization faced by existing methods. This advancement is crucial as it enhances security measures in various applications, ensuring that face recognition technology remains reliable and effective.

Learning Knowledge-based Prompts for Robust 3D Mask Presentation Attack Detection

arXiv:2512.02062v1 Announce Type: cross 
Abstract: Deep learning models are used in safety-critical tasks such as automated driving and face recognition. However, small perturbations in the model input can significantly change the predictions. Adversarial attacks are used to identify small perturbations that can lead to misclassifications. More powerful black-box adversarial attacks are required to develop more effective defenses. A promising approach to black-box adversarial attacks is to repeat the process of extracting a specific image area and changing the perturbations added to it. Existing attacks adopt simple rectangles as the areas where perturbations are changed in a single iteration. We propose applying superpixels instead, which achieve a good balance between color variance and compactness. We also propose a new search method, versatile search, and a novel attack method, Superpixel Attack, which applies superpixels and performs versatile search. Superpixel Attack improves attack success rates by an average of 2.10% compared with existing attacks. Most models used in this study are robust against adversarial attacks, and this improvement is significant for black-box adversarial attacks. The code is avilable at https://github.com/oe1307/SuperpixelAttack.git.

تم اقتراح طريقة جديدة تُدعى Superpixel Attack لتعزيز الهجمات العدائية في نماذج التعلم العميق، خاصة في التطبيقات الحرجة للسلامة مثل القيادة الآلية والتعرف على الوجه. تستخدم هذه الطريقة وحدات بكسل فائقة بدلاً من المستطيلات البسيطة لتطبيق الاضطرابات، مما يحسن فعالية الهجمات العدائية والدفاعات.

Se ha propuesto un nuevo método llamado Superpixel Attack para mejorar los ataques adversariales de caja negra en modelos de aprendizaje profundo, especialmente en aplicaciones críticas para la seguridad como la conducción automatizada y el reconocimiento facial. Este enfoque utiliza superpíxeles en lugar de rectángulos simples para aplicar perturbaciones, mejorando la efectividad de los ataques adversariales y las defensas.

Une nouvelle méthode appelée Superpixel Attack a été proposée pour améliorer les attaques adversariales en boîte noire dans les modèles d'apprentissage profond, en particulier dans des applications critiques pour la sécurité comme la conduite automatisée et la reconnaissance faciale. Cette approche utilise des superpixels au lieu de simples rectangles pour appliquer des perturbations, améliorant ainsi l'efficacité des attaques adversariales et des défenses.

A new method called Superpixel Attack has been proposed to enhance black-box adversarial attacks in deep learning models, particularly in safety-critical applications like automated driving and face recognition. This approach utilizes superpixels instead of simple rectangles to apply perturbations, improving the effectiveness of adversarial attacks and defenses.

Learning Knowledge-based Prompts for Robust 3D Mask Presentation Attack Detection

Was this article worth reading? Share it

WasItAI

Interactive Avatar

Deptho.ai