Research map / Video / Video segmentation and tracking
Celeb-DF: A Large-Scale Challenging Dataset for DeepFake Forensics
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the improved synthesis process to create higher-quality DeepFakes | Celeb-DF compared with existing DeepFake datasets · Celeb-DF and other evaluated datasets | average detection AUC | Hurt |
| Increase synthesized face resolution to 256 × 256 | Celeb-DF synthesis · Celeb-DF synthesized faces | Mask-SSIM | Helped |
| Apply training color augmentation and post-processing color transfer | Celeb-DF synthesis · Celeb-DF synthesized faces | color mismatch | Helped |
| Use a context-inclusive synthesized face and a smoothness mask | Celeb-DF synthesis · Celeb-DF synthesized faces | visible mask boundaries | Helped |
| Filter face-landmark sequences with Kalman smoothing | Celeb-DF synthesis · Celeb-DF videos | temporal flickering | Helped |
| Recompress videos with H.264 c23 instead of using original MPEG4.0 | FWA · Celeb-DF | AUC: 56.9 -> 54.6 | Hurt |
| Recompress videos with H.264 c40 instead of using original MPEG4.0 | FWA · Celeb-DF | AUC: 56.9 -> 52.2 | Hurt |
| Recompress videos with H.264 c23 instead of using original MPEG4.0 | DSP-FWA · Celeb-DF | AUC: 64.6 -> 57.7 | Hurt |
| Recompress videos with H.264 c40 instead of using original MPEG4.0 | DSP-FWA · Celeb-DF | AUC: 64.6 -> 47.2 | Hurt |
| Recompress videos with H.264 c23 instead of using original MPEG4.0 | Xception-c23 · Celeb-DF | AUC: 65.3 -> 65.5 | No clear effect |
| Recompress videos with H.264 c40 instead of using original MPEG4.0 | Xception-c23 · Celeb-DF | AUC: 65.3 -> 52.5 | Hurt |
| Recompress videos with H.264 c23 instead of using original MPEG4.0 | Xception-c40 · Celeb-DF | AUC: 65.5 -> 65.4 | No clear effect |
Closest papers by meaning
- DF40: Toward Next-Generation Deepfake Detection (NeurIPS 2024 · similarity 0.87)
- Generalizing Deepfake Video Detection with Plug-and-Play: Video-Level Blending and Spatiotemporal Adapter Tuning (CVPR 2025 · similarity 0.84)
- Exposing the Deception: Uncovering More Forgery Clues for Deepfake Detection (AAAI 2024 · similarity 0.83)
- Towards a Universal Synthetic Video Detector: From Face or Background Manipulations to Fully AI-Generated Content (CVPR 2025 · similarity 0.83)
- Towards More General Video-based Deepfake Detection through Facial Component Guided Adaptation for Foundation Model (CVPR 2025 · similarity 0.83)
- AVFF: Audio-Visual Feature Fusion for Video Deepfake Detection (CVPR 2024 · similarity 0.83)
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.