Research map / Image generation / Text-to-image diffusion
DynamicID: Zero-Shot Multi-ID Image Personalization with Flexible Facial Editability
Xirui Hu, Jiahao Wang, Hao Chen, Weizhan Zhang, Benqi Wang, Yikun Li, Haishun Nan
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 Semantic-Activated Attention (SAA) (ablation) | DynamicID · Ablation (single-ID, CelebA 500 IDs) | CLIP-T: 0.224 -> 0.239 | Improved |
| add SAA (ablation) | DynamicID · Ablation (single-ID, CelebA 500 IDs) | Expr: 0.422 -> 0.456 | Improved |
| add SAA (ablation) | DynamicID · Ablation (single-ID, CelebA 500 IDs) | Pose: 0.862 -> 0.878 | Improved |
| add SAA (ablation) | DynamicID · Ablation (single-ID, CelebA 500 IDs) | FaceSim: 0.682 -> 0.671 | Hurt |
| add Identity-Motion Reconfigurator (IMR) (ablation) | DynamicID · Ablation (single-ID, CelebA 500 IDs) | Expr: 0.161 -> 0.456 | Improved |
| add IMR (ablation) | DynamicID · Ablation (single-ID, CelebA 500 IDs) | Pose: 0.253 -> 0.878 | Improved |
| add IMR (ablation) | DynamicID · Ablation (single-ID, CelebA 500 IDs) | FaceSim: 0.712 -> 0.671 | Hurt |
| add Direct Feature Matching term (ablation) | IMR training · Ablation (single-ID, CelebA 500 IDs) | FaceSim: 0.663 -> 0.671 | Improved |
| add Direct Feature Matching term (ablation) | IMR training · Ablation (single-ID, CelebA 500 IDs) | Expr: 0.433 -> 0.456 | Improved |
| add Latent Diffusion Consistency term (ablation) | IMR training · Ablation (single-ID, CelebA 500 IDs) | Expr: 0.234 -> 0.456 | Improved |
| add Latent Diffusion Consistency term (ablation) | IMR training · Ablation (single-ID, CelebA 500 IDs) | Pose: 0.644 -> 0.878 | Improved |
| add Latent Diffusion Consistency term (ablation) | IMR training · Ablation (single-ID, CelebA 500 IDs) | FaceSim: 0.667 -> 0.671 | 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.
- PhotoMaker-v2 — PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding (CVPR 2024) · better on 3 of 3 shared results
- MasterWeaver — MasterWeaver: Taming Editability and Face Identity for Personalized Text-to-Image Generation (ECCV 2024) · better on 3 of 3 shared results
- FastComposer — FastComposer: Tuning-Free Multi-subject Image Generation with Localized Attention (IJCV 2024) · better on 2 of 2 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
- FaceA-Net: Facial Attribute-Driven ID Preserving Image Generation Network (AAAI 2025 · similarity 0.89)
- ID-Patch: Robust ID Association for Group Photo Personalization (CVPR 2025 · similarity 0.87)
- DreamSalon: A Staged Diffusion Framework for Preserving Identity-Context in Editable Face Generation (CVPR 2024 · similarity 0.87)
- RealisID: Scale-Robust and Fine-Controllable Identity Customization via Local and Global Complementation (AAAI 2025 · similarity 0.87)
- RectifID: Personalizing Rectified Flow with Anchored Classifier Guidance (NeurIPS 2024 · similarity 0.86)
- PortraitBooth: A Versatile Portrait Model for Fast Identity-preserved Personalization (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.
