Research map / Image generation / Editing and style transfer
StyleSwap: Style-Based Generator Empowers Robust Face Swapping
Zhiliang Xu, Hang Zhou, Zhibin Hong, Ziwei Liu, Jiaming Liu, Zhizhi Guo, Junyu Han, Jingtuo Liu, Errui Ding, Jingdong Wang
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 spatial feature maps instead of a style vector (ablation) | StyleSwap · FF++ | ID Retrieval: 96.68% -> 97.05% | Improved |
| add spatial feature maps instead of a style vector (ablation) | StyleSwap · FF++ | Pose error: 2.81 -> 1.56 | Improved |
| add spatial feature maps instead of a style vector (ablation) | StyleSwap · FF++ | FID: 4.54 -> 2.72 | Improved |
| add the mask branch (ablation) | StyleSwap · FF++ | ID Retrieval: 95.56% -> 97.05% | Improved |
| add the mask branch (ablation) | StyleSwap · FF++ | FID: 3.18 -> 2.72 | Improved |
| add the mask branch (ablation) | StyleSwap · FF++ | Exp. error: 5.78 -> 5.28 | Improved |
| add W-space swapping-driven ID inversion (ablation) | StyleSwap · FF++ | ID Cosine: 0.677 -> 0.693 | Improved |
| use W+ instead of W for ID inversion (ablation) | StyleSwap · FF++ | ID Cosine: 0.693 -> 0.706 | Improved |
| use W+ instead of W for ID inversion (ablation) | StyleSwap · FF++ | FID: 2.60 -> 2.58 | Improved |
| add ID inversion in W+ space (ablation) | StyleSwap · FF++ | Pose error: 1.56 -> 1.51 | No clear effect |
| add 50,000 StyleSwap fakes to detector training | Xception detector · Celeb-DF | AUC: 66.60 -> 73.59 | Improved |
| add 50,000 StyleSwap fakes instead of FaceShifter fakes | Xception detector · Celeb-DF | AUC: 71.92 -> 73.59 | Improved |
Later papers that compared with this method
1 later accepted paper reports StyleSwap as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- DiffSwap: High-Fidelity and Controllable Face Swapping via 3D-Aware Masked Diffusion (CVPR 2023 · better on 4 of 6 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- StyleIPSB: Identity-Preserving Semantic Basis of StyleGAN for High Fidelity Face Swapping (CVPR 2023 · similarity 0.91)
- StyleFace: Towards Identity-Disentangled Face Generation on Megapixels (ECCV 2022 · similarity 0.90)
- Smooth-Swap: A Simple Enhancement for Face-Swapping With Smoothness (CVPR 2022 · similarity 0.89)
- DynamicFace: High-Quality and Consistent Face Swapping for Image and Video using Composable 3D Facial Priors (ICCV 2025 · similarity 0.89)
- Region-Aware Face Swapping (CVPR 2022 · similarity 0.89)
- When StyleGAN Meets Stable Diffusion: a W+ Adapter for Personalized Image Generation (CVPR 2024 · 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.
