Self-Supervised Learning for Detecting AI-Generated Faces as Anomalies
Not on the map. Why?
Abstract (arXiv)
The detection of AI-generated faces is commonly approached as a binary classification task. Nevertheless, the resulting detectors frequently struggle to adapt to novel AI face generators, which evolve rapidly. In this paper, we describe an anomaly detection method for AI-generated faces by leveraging self-supervised learning of camera-intrinsic and face-specific features purely from photographic face images. The success of our method lies in designing a pretext task that trains a feature extractor to rank four ordinal exchangeable image file format (EXIF) tags and classify artificially manipulated face images. Subsequently, we model the learned feature distribution of photographic face images using a Gaussian mixture model. Faces with low likelihoods are flagged as AI-generated. Both quantitative and qualitative experiments validate the effectiveness of our method. Our code is available at urlhttps://github.com/MZMMSEC/AIGFDEXIF.git.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 36 of its 52 references that are in the map. First posted on arXiv 4 Jan 2025.
Builds on
36 of 52 references.
References · 36
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
- 2020Denoising Diffusion Implicit Modelscited by 9,655
- 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
- 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
- 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
- 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
- 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
- 2016Learning without Forgettingcited by 2,465
- 2019CNN-generated images are surprisingly easy to spot... for nowcited by 842
24 more
- 2017How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)cited by 524
- 2020Leveraging Frequency Analysis for Deep Fake Image Recognitioncited by 501
- 2023Towards Universal Fake Image Detectors that Generalize Across Generative Modelscited by 483
- 2023Aligning Text-to-Image Models using Human Feedbackcited by 418
- 2023DIRE for Diffusion-Generated Image Detectioncited by 386
- 2018Exposing Deep Fakes Using Inconsistent Head Posescited by 294
- 2023FreeDoM: Training-Free Energy-Guided Conditional Diffusion Modelcited by 290
- 2022On the detection of synthetic images generated by diffusion modelscited by 282
- 2020Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral Distributionscited by 278
- 2021Generalizing Face Forgery Detection with High-frequency Featurescited by 261
- 2020Global Texture Enhancement for Fake Face Detection in the Wildcited by 220
- 2021General Facial Representation Learning in a Visual-Linguistic Mannercited by 193
- 2021Are GAN generated images easy to detect? A critical analysis of the state-of-the-artcited by 139
- 2018Fighting Fake News: Image Splice Detection via Learned Self-Consistencycited by 136
- 2022Leveraging Real Talking Faces via Self-Supervision for Robust Forgery Detectioncited by 130
- 2023Intriguing properties of synthetic images: from generative adversarial networks to diffusion modelscited by 121
- 2019DeepPrivacy: A Generative Adversarial Network for Face Anonymizationcited by 108
- 2020Exposing GAN-generated Faces Using Inconsistent Corneal Specular Highlightscited by 59
- 2024Zero-Shot Detection of AI-Generated Imagescited by 50
- 2021Eyes Tell All: Irregular Pupil Shapes Reveal GAN-generated Facescited by 48
- 2024Diffusion Facial Forgery Detectioncited by 41
- 2024DiffusionFace: Towards a Comprehensive Dataset for Diffusion-Based Face Forgery Analysiscited by 29
- 2023EXIF as Language: Learning Cross-Modal Associations Between Images and Camera Metadatacited by 17
- 2024Semantics-Oriented Multitask Learning for DeepFake Detection: A Joint Embedding Approachcited by 4
Built on by
Citing papers per year: 2025: 2; 2026: 1.
3 papers.
- 2025Bi-Level Optimization for Self-Supervised AI-Generated Face Detectioncited by 4
- 2025DGS-Net: Distillation-Guided Gradient Surgery for CLIP Fine-Tuning in AI-Generated Image Detectioncited by 4
- 2026Color Matters: Demosaicing-Guided Color Correlation Training for Generalizable AI-Generated Image Detectioncited by 0
- 2026Color Matters: Demosaicing-Guided Color Correlation Training for Generalizable AI-Generated Image Detection30 Jan 2026
- 2025DGS-Net: Distillation-Guided Gradient Surgery for CLIP Fine-Tuning in AI-Generated Image Detection17 Nov 2025
- 2025Bi-Level Optimization for Self-Supervised AI-Generated Face Detection30 Jul 2025