Research map / Video / Video segmentation and tracking
Improving the Efficiency and Robustness of Deepfakes Detection Through Precise Geometric Features
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 geometric landmarks with temporal RNN modeling instead of image-appearance features | LRNet · FF++ | AUC (%): 99.7 -> 99.9 | Helped |
| use LRNet rather than appearance-based Xception | FF++ c40 · FF++ c40 | AUC (%): 86.5 -> 95.7 | Helped |
| use LRNet rather than the strongest listed baseline | Celeb-DF · Celeb-DF | AUC (%): 64.6 -> 56.9 | Hurt |
| use LRNet trained on FF++ raw when testing compressed videos | LRNet · FF++ c40 | AUC (%): Xception: 99.7 -> 86.5; LRNet: 99.9 -> 95.7 | Helped |
| use LRNet under noise corruption rather than the strongest listed noisy baseline | DF1.0 std/noise · DF1.0 std/noise | accuracy (%): 91.50 -> 96.83 | Helped |
| add the Kalman filter to landmark calibration (ablation) | LRNet · FF++ raw | accuracy (%): 98.5 -> 99.7 | Helped |
| add the Kalman filter to landmark calibration (ablation) | LRNet · FF++ c40 | AUC (%): 94.3 -> 95.7 | Helped |
| add landmark calibration (ablation) | LRNet · FF++ raw | accuracy (%): 92.8 -> 99.7 | Helped |
| add landmark calibration (ablation) | LRNet · FF++ c40 | AUC (%): 92.6 -> 95.7 | Helped |
| add the g2 stream for landmark-difference patterns (ablation) | LRNet · FF++ raw | accuracy (%): 83.4 -> 99.7 | Helped |
| add the g1 stream for landmark-shape movement patterns (ablation) | LRNet · FF++ c40 | AUC (%): 93.9 -> 95.7 | Helped |
Closest papers by meaning
- Exploiting Style Latent Flows for Generalizing Deepfake Video Detection (CVPR 2024 · similarity 0.88)
- Generalizing Deepfake Video Detection with Plug-and-Play: Video-Level Blending and Spatiotemporal Adapter Tuning (CVPR 2025 · similarity 0.88)
- Exposing the Deception: Uncovering More Forgery Clues for Deepfake Detection (AAAI 2024 · similarity 0.86)
- Standing on the Shoulders of Giants: Reprogramming Visual-Language Model for General Deepfake Detection (AAAI 2025 · similarity 0.86)
- Towards More General Video-based Deepfake Detection through Facial Component Guided Adaptation for Foundation Model (CVPR 2025 · similarity 0.85)
- DF40: Toward Next-Generation Deepfake Detection (NeurIPS 2024 · similarity 0.85)
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.