Research map / Video / Video segmentation and tracking
Seeing Through Deepfakes: A Human-Inspired Framework for Multi-Face Detection
Juan Hu, Shaojing Fan, Terence Sim
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 M2 (multi-face classification and contrastive loss) to M1 (ablation) | HICOM · FFIW | FCAC: 87.7 -> 89.9 | Improved |
| add M2 to M1 (ablation) | HICOM · OpenForensics | FCAC: 93.9 -> 95.7 | Improved |
| add M2 to M1 (ablation) | HICOM · DF-Platter | FCAC: 90.4 -> 92.0 | Improved |
| add M3 (gaze alignment) to M1+M2 (ablation) | HICOM · FFIW | FCAC: 89.9 -> 90.6 | Improved |
| add M3 to M1+M2 (ablation) | HICOM · OpenForensics | FCAC: 95.7 -> 97.2 | Improved |
| add M3 to M1+M2 (ablation) | HICOM · DF-Platter | FCAC: 92.0 -> 93.3 | Improved |
| add M4 (face-body consistency) to M1+M2+M3 (ablation) | HICOM · FFIW | FCAC: 90.6 -> 91.3 | Improved |
| add M4 to M1+M2+M3 (ablation) | HICOM · OpenForensics | FCAC: 97.2 -> 97.8 | Improved |
| add M4 to M1+M2+M3 (ablation) | HICOM · DF-Platter | FCAC: 93.3 -> 93.5 | Improved |
| add M4 to M1+M2+M3 (ablation) | HICOM · DF-Platter | FCAU: 94.0 -> 94.6 | Improved |
| add M2, M3 and M4 to M1 (ablation) | HICOM · FFIW | FCAU: 89.0 -> 92.1 | Improved |
| use human-inspired multi-cue HICOM instead of best prior method under perturbation | HICOM vs TALL · OpenForensics with perturbations | FCAC: 77.1 -> 78.6 | Improved |
Closest papers by meaning
- Guard Me If You Know Me: Protecting Specific Face-Identity from Deepfakes (NeurIPS 2025 · similarity 0.89)
- From Specificity to Generality: Revisiting Generalizable Artifacts in Detecting Face Deepfakes (NeurIPS 2025 · similarity 0.88)
- DF40: Toward Next-Generation Deepfake Detection (NeurIPS 2024 · similarity 0.88)
- Towards a Universal Synthetic Video Detector: From Face or Background Manipulations to Fully AI-Generated Content (CVPR 2025 · similarity 0.88)
- TriDF: Evaluating Perception, Detection, and Hallucination for Interpretable DeepFake Detection (CVPR 2026 · similarity 0.88)
- DiffusionFake: Enhancing Generalization in Deepfake Detection via Guided Stable Diffusion (NeurIPS 2024 · 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.
