Research map / Image generation / Editing and style transfer
FlowFace: Rectifying Identity Conditioning with Riemannian Geometry for Face Generation
Xinran Deng, Yiling Wu, Ye Tian, Libo Zhang
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 dual-stream manifold encoder (ablation) | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | SimTarget: 48.3 -> 64.8 | Improved |
| add Lie-bracket (BCH) fusion (ablation) | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | SimTarget: 52.1 -> 64.8 | Improved |
| add geodesic regularization (ablation) | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | SimTarget: 55.7 -> 64.8 | Improved |
| add directional covariant regularization (ablation) | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | SimTarget: 58.4 -> 64.8 | Improved |
| add dual-stream manifold encoder (ablation) | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | FID: 95.4 -> 76.8 | Improved |
| add Lie-bracket fusion (ablation) | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | Expr.Acc: 42.6 -> 52.4 | Improved |
| add BCH interaction term to composition (ablation) | FlowFace · Two-edit composition | OrderGap-ID: 0.146 -> 0.052 | Improved |
| add BCH interaction term (ablation) | FlowFace · Three-edit composition | Avg. PermGap-ID: 0.182 -> 0.071 | Improved |
| BCH fusion instead of vector addition | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | SimTarget: 56.4 -> 64.8 | Improved |
| BCH fusion instead of attention weighting | FlowFace · FFHQ/CelebA-HQ identity-cluster benchmark | SimTarget: 61.6 -> 64.8 | Improved |
| learned SPD metric training instead of Euclidean interpolation | FlowFace · Trajectory stability | Collapse Rate: 12.3% -> 1.2% | Improved |
| learned SPD metric instead of hypersphere projection | FlowFace · Trajectory stability | Smoothness: 0.89 -> 0.96 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- HiFi-Portrait — HiFi-Portrait: Zero-shot Identity-preserved Portrait Generation with High-fidelity Multi-face Fusion (CVPR 2025) · better on 4 of 4 shared results
- PhotoMaker — PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding (CVPR 2024) · better on 3 of 3 shared results
- FastComposer — FastComposer: Tuning-Free Multi-subject Image Generation with Localized Attention (IJCV 2024) · better on 3 of 3 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
- FlowFace: Semantic Flow-Guided Shape-Aware Face Swapping (AAAI 2023 · similarity 0.88)
- When StyleGAN Meets Stable Diffusion: a W+ Adapter for Personalized Image Generation (CVPR 2024 · similarity 0.88)
- ID-to-3D: Expressive ID-guided 3D Heads via Score Distillation Sampling (NeurIPS 2024 · similarity 0.87)
- Data Synthesis with Diverse Styles for Face Recognition via 3DMM-Guided Diffusion (CVPR 2025 · similarity 0.86)
- Text-Guided 3D Face Synthesis - From Generation to Editing (CVPR 2024 · similarity 0.86)
- $\text{ID}^3$: Identity-Preserving-yet-Diversified Diffusion Models for Synthetic Face Recognition (NeurIPS 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.
