Research map / Detection and segmentation / Open-set and few-shot detection
Advancing High Fidelity Identity Swapping for Forgery Detection
Lingzhi Li, Jianmin Bao, Hao Yang, Dong Chen, Fang Wen
Check 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 |
|---|---|---|---|
| Adding attentional mask fusion (AAD with mask M^k) versus plain element-wise addition (Add) in the AAD layers (ablation) | AEI-Net · qualitative comparison (Figure 7) | visual quality (blur/ghosting) | Improved |
| Using multi-level attribute embeddings instead of only the first three levels (Compressed) (ablation) | AEI-Net · qualitative comparison (Figure 7) | visual quality (blurriness) | Improved |
| Adding the second-stage HEAR-Net refinement to AEI-Net (ablation) | FaceShifter · qualitative (Figure 10) | occlusion preservation and color correction | Improved |
| Replacing pixel-level replacement/3D pipelines with FaceShifter's subject-agnostic GAN approach (identity-preserving swap) | FaceShifter · FaceForensics++ videos | ID retrieval: 97.38 vs 82.41 for next best (IPGAN) | Improved |
| FaceShifter vs. baselines on expression preservation | FaceShifter · FaceForensics++ videos | expression error: 2.06 vs 2.14 for best baseline (FaceSwap) | Improved |
| FaceShifter vs. FaceSwap on pose error | FaceShifter · FaceForensics++ videos | pose error: 2.96 vs 2.51 for FaceSwap | No clear effect |
| Face X-Ray forgery detector vs. FF++ detector on FaceShifter outputs | Forgery detection · 5000 swapped + 5000 real images | AUC: 96.82 vs 52.22 | Improved |
Closest papers by meaning
- DF40: Toward Next-Generation Deepfake Detection (NeurIPS 2024 · similarity 0.89)
- DiffusionFake: Enhancing Generalization in Deepfake Detection via Guided Stable Diffusion (NeurIPS 2024 · similarity 0.88)
- From Specificity to Generality: Revisiting Generalizable Artifacts in Detecting Face Deepfakes (NeurIPS 2025 · similarity 0.87)
- FakeDiffer: Distributional Disparity Learning on Differentiated Reconstruction for Face Forgery Detection (AAAI 2025 · similarity 0.86)
- FreqBlender: Enhancing DeepFake Detection by Blending Frequency Knowledge (NeurIPS 2024 · similarity 0.86)
- Critical Forgetting-Based Multi-Scale Disentanglement for Deepfake Detection (AAAI 2025 · similarity 0.86)
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.
