mlsanityData to 7 Oct 2026

Exposing GAN-generated Faces Using Inconsistent Corneal Specular Highlights

arXiv 2009.11924 ↗First posted 24 Sep 2020cs.CVcited by 59

Not on the map. Why?

Abstract (arXiv)

Sophisticated generative adversary network (GAN) models are now able to synthesize highly realistic human faces that are difficult to discern from real ones visually. In this work, we show that GAN synthesized faces can be exposed with the inconsistent corneal specular highlights between two eyes. The inconsistency is caused by the lack of physical/physiological constraints in the GAN models. We show that such artifacts exist widely in high-quality GAN synthesized faces and further describe an automatic method to extract and compare corneal specular highlights from two eyes. Qualitative and quantitative evaluations of our method suggest its simplicity and effectiveness in distinguishing GAN synthesized faces.

Key facts. Cited by 59 papers in the map (data through 7 Oct 2026). First posted on arXiv 24 Sep 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 9; 2023: 10; 2024: 14; 2025: 14; 2026: 7.

59 papers.

Sort:

By area:

  1. 2024AEROBLADE: Training-Free Detection of Latent Diffusion Images Using Autoencoder Reconstruction Errorcited by 123
  2. 2022Towards the Detection of Diffusion Model Deepfakescited by 111
  3. 2024Preserving Fairness Generalization in Deepfake Detectioncited by 70
  4. 2024Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensicscited by 68
  5. 2021Countering Malicious DeepFakes: Survey, Battleground, and Horizoncited by 56
  6. 2021Eyes Tell All: Irregular Pupil Shapes Reveal GAN-generated Facescited by 48
  7. 2023Improving Fairness in Deepfake Detectioncited by 40
  8. 2023PTW: Pivotal Tuning Watermarking for Pre-Trained Image Generatorscited by 34
  9. 2022GAN-generated Faces Detection: A Survey and New Perspectivescited by 29
  10. 2021Robust Attentive Deep Neural Network for Exposing GAN-generated Facescited by 28
  11. 2024Robust CLIP-Based Detector for Exposing Diffusion Model-Generated Imagescited by 20
  12. 2024UnMarker: A Universal Attack on Defensive Image Watermarkingcited by 19
38 more
  1. 2024Characteristics and prevalence of fake social media profiles with AI-generated facescited by 14
  2. 2025From Specificity to Generality: Revisiting Generalizable Artifacts in Detecting Face Deepfakescited by 14
  3. 2025Characterizing Photorealism and Artifacts in Diffusion Model-Generated Imagescited by 13
  4. 2023Synthesizing Black-box Anti-forensics DeepFakes with High Visual Qualitycited by 12
  5. 2022Open-Eye: An Open Platform to Study Human Performance on Identifying AI-Synthesized Facescited by 11
  6. 2022PseudoProp: Robust Pseudo-Label Generation for Semi-Supervised Object Detection in Autonomous Driving Systemscited by 10
  7. 2024AI-Generated Faces in the Real World: A Large-Scale Case Study of Twitter Profile Imagescited by 10
  8. 2021Where Do Deep Fakes Look? Synthetic Face Detection via Gaze Trackingcited by 9
  9. 2024Masked Conditional Diffusion Model for Enhancing Deepfake Detectioncited by 9
  10. 2023Finding AI-Generated Faces in the Wildcited by 8
  11. 2024Semantic Contextualization of Face Forgery: A New Definition, Dataset, and Detection Methodcited by 8
  12. 2022An Overview of Recent Work in Media Forensics: Methods and Threatscited by 7
  13. 2023Harnessing the Power of Text-image Contrastive Models for Automatic Detection of Online Misinformationcited by 6
  14. 2024Adversarial Robustness of AI-Generated Image Detectors in the Real Worldcited by 6
  15. 2022How Do Deepfakes Move? Motion Magnification for Deepfake Source Detectioncited by 5
  16. 2023Towards Robust GAN-generated Image Detection: a Multi-view Completion Representationcited by 5
  17. 2023Exploiting Facial Relationships and Feature Aggregation for Multi-Face Forgery Detectioncited by 5
  18. 2025Robust AI-Generated Face Detection with Imbalanced Datacited by 5
  19. 2021Distinguishing Natural and Computer-Generated Images using Multi-Colorspace fused EfficientNetcited by 4
  20. 2022Aggregating Layers for Deepfake Detectioncited by 4
  21. 2024Semantics-Oriented Multitask Learning for DeepFake Detection: A Joint Embedding Approachcited by 4
  22. 2024Passive Deepfake Detection Across Multi-modalities: A Comprehensive Surveycited by 4
  23. 2025Rethinking Individual Fairness in Deepfake Detectioncited by 4
  24. 2025Bi-Level Optimization for Self-Supervised AI-Generated Face Detectioncited by 4
  25. 2023Pneumonia Detection in Chest X-Ray Images : Handling Class Imbalancecited by 3
  26. 2023Feature Extraction Matters More: Universal Deepfake Disruption through Attacking Ensemble Feature Extractorscited by 3
  27. 2024Transfer Learning of Real Image Features with Soft Contrastive Loss for Fake Image Detectioncited by 3
  28. 2025Self-Supervised Learning for Detecting AI-Generated Faces as Anomaliescited by 3
  29. 2025Unmasking Puppeteers: Leveraging Biometric Leakage to Expose Impersonation in AI-Based Videoconferencingcited by 2
  30. 2026Can We Build a Monolithic Model for Fake Image Detection? SICA: Semantic-Induced Constrained Adaptation for Unified-Yet-Discriminative Artifact Feature Space Reconstructioncited by 2
  31. 2022An Architecture for the detection of GAN-generated Flood Images with Localization Capabilitiescited by 1
  32. 2022Detection of Real-time DeepFakes in Video Conferencing with Active Probing and Corneal Reflectioncited by 1
  33. 2023Spritz-PS: Validation of Synthetic Face Images Using a Large Dataset of Printed Documentscited by 1
  34. 2025Combating Falsification of Speech Videos with Live Optical Signatures (Extended Version)cited by 1
  35. 2025Is It Certainly a Deepfake? Reliability Analysis in Detection & Generation Ecosystemcited by 1
  36. 2025A Dual-Branch CNN for Robust Detection of AI-Generated Facial Forgeriescited by 1
  37. 2025UMCL: Unimodal-generated Multimodal Contrastive Learning for Cross-compression-rate Deepfake Detectioncited by 1
  38. 2025Frequency Bias Matters: Diving into Robust and Generalized Deep Image Forgery Detectioncited by 1

50 most cited of 59 All 59 (JSON)