Research map / Video / Video segmentation and tracking
ID-Reveal: Identity-Aware DeepFake Video Detection
Davide Cozzolino, Andreas Rössler, Justus Thies, Matthias Nießner, Luisa Verdoliva
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add adversarial training with 3DMM generative network (ablation) | ID-Reveal · DFD FR HQ | accuracy (%): 73.8 -> 75.6 | Improved |
| add adversarial training with 3DMM generative network (ablation) | ID-Reveal · DFD FR LQ | accuracy (%): 79.1 -> 81.8 | Improved |
| add adversarial training with 3DMM generative network (ablation) | ID-Reveal · DFD FS HQ | accuracy (%): 82.6 -> 84.8 | Improved |
| add adversarial training with 3DMM generative network (ablation) | ID-Reveal · DFD FS LQ | accuracy (%): 73.0 -> 78.1 | Improved |
| add adversarial training (ablation) | ID-Reveal · DFD FR (hardest scenario) | AUC: further improvement of around 3% for AUC | Improved |
| use proposed loss instead of multi-similarity loss (ablation) | ID-Reveal · DFD (HQ/LQ, FR/FS) | AUC: 5.5% on average | Improved |
| use proposed loss instead of triplet loss (ablation) | ID-Reveal · DFD (HQ/LQ, FR/FS) | AUC: 2.8% on average | Improved |
| use proposed loss instead of multi-similarity loss, with adversarial (ablation) | ID-Reveal · DFD FR LQ | accuracy (%): 69.1 -> 81.8 | Improved |
| average frame scores instead of taking maximum (ablation) | frame-based detectors · DFD | AUC: increase in terms of AUC is around 0.04, accuracy increases of about 3% | Improved |
| move from HQ to LQ compression | Xception · DFD FS | accuracy (%): 78.5 -> 58.5 | Hurt |
| move from HQ to LQ compression | ID-Reveal · DFD FS | accuracy (%): 84.8 -> 78.1 | Hurt |
| train only on real videos (VoxCeleb2) instead of supervised on DFDC | ID-Reveal vs Seferbekov · DFD FR HQ | accuracy (%): 55.8 -> 75.6 | Improved |
Later papers that compared with this method
1 later accepted paper reports ID-Reveal as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Avatar Fingerprinting for Authorized Use of Synthetic Talking-Head Videos (ECCV 2024 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Exposing the Deception: Uncovering More Forgery Clues for Deepfake Detection (AAAI 2024 · similarity 0.90)
- Identity-Aware Vision-Language Model for Explainable Face Forgery Detection (AAAI 2026 · similarity 0.89)
- Generalizing Deepfake Video Detection with Plug-and-Play: Video-Level Blending and Spatiotemporal Adapter Tuning (CVPR 2025 · similarity 0.89)
- DF40: Toward Next-Generation Deepfake Detection (NeurIPS 2024 · similarity 0.89)
- Guard Me If You Know Me: Protecting Specific Face-Identity from Deepfakes (NeurIPS 2025 · similarity 0.89)
- Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector (CVPR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
