Research map / Image generation / Editing and style transfer
One Shot Face Swapping on Megapixels
Yuhao Zhu, Qi Li, Jian Wang, Cheng-Zhong Xu, Zhenan Sun
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 |
|---|---|---|---|
| Use W++ instead of W+ for face reconstruction (ablation) | HieRFE · CelebA-HQ face reconstruction | LPIPS: 0.2486 -> 0.2335 | Improved |
| Use W++ instead of W+ for face reconstruction (ablation) | HieRFE · CelebA-HQ face reconstruction | MSE: 0.0672 -> 0.0563 | Improved |
| Use W++ instead of W+ for face reconstruction (ablation) | HieRFE · CelebA-HQ face reconstruction | failure rate: 1.28% -> 0.65% | Improved |
| Use W++ instead of W+ for face swapping (ablation) | MegaFS · CelebA-HQ face swapping | ID similarity: 0.5438 -> 0.5816 | Improved |
| Use W++ instead of W+ for face swapping (ablation) | MegaFS · CelebA-HQ face swapping | pose: 4.0640 -> 3.8179 | Improved |
| Use W++ instead of W+ for face swapping (ablation) | MegaFS · CelebA-HQ face swapping | expression: 1.7467 -> 1.6489 | Improved |
| Use FTM instead of Latent Code Replacement (ablation) | MegaFS · CelebA-HQ face swapping | ID similarity: 0.3997 -> 0.5014 | Improved |
| Use FTM instead of Latent Code Replacement (ablation) | MegaFS · CelebA-HQ face swapping | pose: 5.04 -> 3.58 | Improved |
| Use FTM instead of Latent Code Replacement (ablation) | MegaFS · CelebA-HQ face swapping | FID: 9.64 -> 10.16 | Hurt |
| Use FTM instead of ID Injection (ablation) | MegaFS · CelebA-HQ face swapping | ID similarity: 0.4447 -> 0.5014 | Improved |
| Use FTM instead of ID Injection (ablation) | MegaFS · CelebA-HQ face swapping | pose: 3.67 -> 3.58 | Improved |
| Use FTM instead of ID Injection (ablation) | MegaFS · CelebA-HQ face swapping | expression: 2.82 -> 2.87 | Hurt |
Closest papers by meaning
- MFIM: Megapixel Facial Identity Manipulation (ECCV 2022 · similarity 0.90)
- DeepFake Detection Based on Discrepancies Between Faces and Their Context (TPAMI 2021 · similarity 0.88)
- DF40: Toward Next-Generation Deepfake Detection (NeurIPS 2024 · similarity 0.86)
- Zero-Shot Head Swapping in Real-World Scenarios (CVPR 2025 · similarity 0.85)
- DiffusionFake: Enhancing Generalization in Deepfake Detection via Guided Stable Diffusion (NeurIPS 2024 · similarity 0.85)
- From Specificity to Generality: Revisiting Generalizable Artifacts in Detecting Face Deepfakes (NeurIPS 2025 · 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.
