FaceCat: Enhancing Face Recognition Security with a Unified Diffusion Model
On the map: Diffusion models → Robustness, privacy and adversarial aspects
Abstract (arXiv)
Face anti-spoofing (FAS) and adversarial detection (FAD) have been regarded as critical technologies to ensure the safety of face recognition systems. However, due to limited practicality, complex deployment, and the additional computational overhead, it is necessary to implement both detection techniques within a unified framework. This paper aims to achieve this goal by breaking through two primary obstacles: 1) the suboptimal face feature representation and 2) the scarcity of training data. To address the limited performance caused by existing feature representations, motivated by the rich structural and detailed features of face diffusion models, we propose FaceCat, the first approach leveraging the diffusion model to simultaneously enhance the performance of FAS and FAD. Specifically, FaceCat elaborately designs a hierarchical fusion mechanism to capture rich face semantic features of the diffusion model. These features then serve as a robust foundation for a lightweight head, designed to execute FAS and FAD simultaneously. Due to the limitations in feature representation that arise from relying solely on single-modality image data, we further propose a novel text-guided multi-modal alignment strategy that utilizes text prompts to enrich feature representation, thereby enhancing performance. To combat data scarcity, we build a comprehensive dataset with a wide range of 28 attack types, offering greater potential for a unified framework in facial security. Extensive experiments validate the effectiveness of FaceCat generalizes significantly better and obtains excellent robustness against common input transformations.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 37 of its 43 references that are in the map. First posted on arXiv 14 Apr 2024.
Builds on
37 of 43 references.
References · 37
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
- 2020Denoising Diffusion Probabilistic Modelscited by 20,851
- 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,845
- 2016Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learningcited by 1,509
- 2017In Defense of the Triplet Loss for Person Re-Identificationcited by 784
- 2021DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulationcited by 640
- 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
25 more
- 2014Semi-Supervised Learning with Deep Generative Modelscited by 494
- 2016Adversarial Feature Learningcited by 350
- 2019Large Scale Adversarial Representation Learningcited by 238
- 2016A study of the effect of JPG compression on adversarial imagescited by 196
- 2019Efficient Decision-based Black-box Adversarial Attacks on Face Recognitioncited by 173
- 2020Searching Central Difference Convolutional Networks for Face Anti-Spoofingcited by 173
- 2018Learning Deep Models for Face Anti-Spoofing: Binary or Auxiliary Supervisioncited by 167
- 2019AdvHat: Real-world adversarial attack on ArcFace Face ID systemcited by 161
- 2019Biometric Face Presentation Attack Detection with Multi-Channel Convolutional Neural Networkcited by 78
- 2019Deep Pixel-wise Binary Supervision for Face Presentation Attack Detectioncited by 76
- 2019Deep Tree Learning for Zero-shot Face Anti-Spoofingcited by 73
- 2023FLIP: Cross-domain Face Anti-spoofing with Language Guidancecited by 45
- 2021LiBRe: A Practical Bayesian Approach to Adversarial Detectioncited by 40
- 2023Towards Effective Adversarial Textured 3D Meshes on Physical Face Recognitioncited by 33
- 2021Unified Detection of Digital and Physical Face Attackscited by 31
- 2020Deep Models and Shortwave Infrared Information to Detect Face Presentation Attackscited by 29
- 2022Benchmarking Joint Face Spoofing and Forgery Detection with Visual and Physiological Cuescited by 24
- 2022Flexible-Modal Face Anti-Spoofing: A Benchmarkcited by 20
- 20213D High-Fidelity Mask Face Presentation Attack Detection Challengecited by 18
- 2019Detection of Face Recognition Adversarial Attackscited by 15
- 2020FaceGuard: A Self-Supervised Defense Against Adversarial Face Imagescited by 15
- 2021FACESEC: A Fine-grained Robustness Evaluation Framework for Face Recognition Systemscited by 14
- 2021FRT-PAD: Effective Presentation Attack Detection Driven by Face Related Taskcited by 7
- 2023AdvFAS: A robust face anti-spoofing framework against adversarial examplescited by 3
- 2022Adversarial Detection without Model Informationcited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025FaceShield: Explainable Face Anti-Spoofing with Multimodal Large Language Models
- 2024DiffFAS: Face Anti-Spoofing via Generative Diffusion Models
- 2025SLIP: Spoof-Aware One-Class Face Anti-Spoofing with Language Image Pretraining
- 2023DiffProtect: Generate Adversarial Examples with Diffusion Models for Facial Privacy Protection
- 2023DiFace: Cross-Modal Face Recognition through Controlled Diffusion
- 2026Adv-TGD: Adversarial Text-Guided Diffusion for Face Recognition Impersonation Attacks
- 2026From Intuition to Investigation: A Tool-Augmented Reasoning MLLM Framework for Generalizable Face Anti-Spoofing
- 2026Benchmarking Vision Foundation Models for Domain-Generalizable Face Anti-Spoofing