# Facial Forgery Detection on FaceForensics++ Hidden Benchmark low quality (test) Source: [Facial Forgery Detection on FaceForensics++ Hidden Benchmark low quality (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:58:47.856Z Evaluation of Facial Forgery Detection performance on the low-quality test split of the FaceForensics++ dataset, focusing on precision across specific forgery types (DF, F2F, FS, NT, Real) and overall accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Facial Forgery Detection | | Dataset (including any stated split) | FaceForensics++ Hidden Benchmark low quality (test) | | Metric | Precision (DF) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 7 of 7 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Precision (DF)](): Higher is better; unit: PERCENT. - [Precision (F2F)](): Higher is better; unit: PERCENT. - [Precision (FS)](): Higher is better; unit: PERCENT. - [Precision (NT)](): Higher is better; unit: PERCENT. - [Precision (Real)](): Higher is better; unit: PERCENT. - [Total Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Precision (DF) | Result date | Report paper | | --- | --- | --- | --- | | XceptionNet | 96.36 | Not recorded | 68fa66421e209131cea0f418 | | MesoNet | 87.27 | Not recorded | 68fa66421e209131cea0f418 | | Cozzolino et al. | 85.45 | Not recorded | 68fa66421e209131cea0f418 | | Rahmouni et al. | 85.45 | Not recorded | 68fa66421e209131cea0f418 | | Bayar and Stamm | 84.55 | Not recorded | 68fa66421e209131cea0f418 | | Xcept. Full Image | 74.55 | Not recorded | 68fa66421e209131cea0f418 | | Steg. Features | 73.64 | Not recorded | 68fa66421e209131cea0f418 | ### XceptionNet Precision (DF): 96.36. Result ID: 698ce6498b517eb4331d8926. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Region":"Face Region"}. This record does not assert independent reproduction. ### MesoNet Precision (DF): 87.27. Result ID: 698ce6498b517eb4331d8925. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input Region":"Face Region"}. This record does not assert independent reproduction. ### Cozzolino et al. Precision (DF): 85.45. Result ID: 698ce6498b517eb4331d8922. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input Region":"Face Region"}. This record does not assert independent reproduction. ### Rahmouni et al. Precision (DF): 85.45. Result ID: 698ce6498b517eb4331d8923. [Reporting paper]() Recorded evaluation configuration: {"Input Region":"Face Region"}. This record does not assert independent reproduction. ### Bayar and Stamm Precision (DF): 84.55. Result ID: 698ce6498b517eb4331d8924. [Reporting paper]() Recorded evaluation configuration: {"Input Region":"Face Region"}. This record does not assert independent reproduction. ### Xcept. Full Image Precision (DF): 74.55. Result ID: 698ce6498b517eb4331d8920. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input Region":"Full Image"}. This record does not assert independent reproduction. ### Steg. Features Precision (DF): 73.64. Result ID: 698ce6498b517eb4331d8921. [Reporting paper]() Recorded evaluation configuration: {"Input Region":"Face Region"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.