Research map / Video / Video segmentation and tracking
Learning Natural Consistency Representation for Face Forgery Video Detection
Daichi Zhang, Zihao Xiao, Shikun Li, Fanzhao Lin, Jianmin Li, Shiming Ge
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 self-supervised natural consistency pretraining on real videos (ablation) | NACO · Celeb-DF (train FF++) | video-level AUC (%): 67.9 -> 89.5 | Improved |
| add self-supervised pretraining (ablation) | NACO · Unseen manipulations FF++ average | video-level AUC (%): 93.6 -> 99.7 | Improved |
| add Temporal Contrastive Module to SPM (ablation) | NACO · Celeb-DF | AUC (%): 80.6 -> 89.5 | Improved |
| add Spatial Predictive Module to TCM (ablation) | NACO · Celeb-DF | AUC (%): 79.2 -> 89.5 | Improved |
| add both SPM and TCM versus SPM only (ablation) | NACO · FF++ (c40) | AUC (%): 94.6 -> 95.4 | Improved |
| increase the number of real videos used in pretraining (ablation) | NACO · FF++ (c40), Celeb-DF | AUC | Improved |
| NACO versus RealForensics on FaceShifter | NACO · FaceShifter (train FF++) | video-level AUC (%): 99.7 -> 99.4 | Hurt |
| NACO versus RealForensics on average perturbation robustness | NACO · FF++ perturbations | average AUC (%): 95.6 -> 96.2 | Improved |
| NACO compared with FTCN on unseen manipulations | NACO · FF++ unseen manipulations average | video-level AUC (%): 99.7 vs 99.7 | No clear effect |
Closest papers by meaning
- Delving Deep Into Unraveling Spatio-Temporal Forgery Cues for Face Forgery Video Detection (TPAMI 2026 · similarity 0.91)
- Face Forgery Video Detection via Temporal Forgery Cue Unraveling (CVPR 2025 · similarity 0.88)
- Identity-Aware Vision-Language Model for Explainable Face Forgery Detection (AAAI 2026 · similarity 0.88)
- Generalizing Deepfake Video Detection with Plug-and-Play: Video-Level Blending and Spatiotemporal Adapter Tuning (CVPR 2025 · similarity 0.87)
- FakeDiffer: Distributional Disparity Learning on Differentiated Reconstruction for Face Forgery Detection (AAAI 2025 · similarity 0.87)
- Stacking Brick by Brick: Aligned Feature Isolation for Incremental Face Forgery Detection (CVPR 2025 · similarity 0.87)
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.
