Research map / Image generation / Text-to-image diffusion
MasterWeaver: Taming Editability and Face Identity for Personalized Text-to-Image Generation
Yuxiang Wei, Zhilong Ji, Jinfeng Bai, Hongzhi Zhang, Lei Zhang, Wangmeng Zuo
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 editing direction loss L_edit (ablation) | MasterWeaver · CelebA-HQ single reference | CLIP-T: 0.209 -> 0.232 | Improved |
| add editing direction loss L_edit (ablation) | MasterWeaver · CelebA-HQ single reference | Face similarity: 0.637 -> 0.631 | Hurt |
| add background disentanglement loss L_disen (ablation) | MasterWeaver · CelebA-HQ single reference | CLIP-T: 0.226 -> 0.232 | Improved |
| add background disentanglement loss L_disen (ablation) | MasterWeaver · CelebA-HQ single reference | Face similarity: 0.617 -> 0.631 | Improved |
| add face-augmented dataset (ablation) | MasterWeaver · CelebA-HQ single reference | CLIP-T: 0.221 -> 0.232 | Improved |
| add face-augmented dataset (ablation) | MasterWeaver · CelebA-HQ single reference | Face similarity: 0.634 -> 0.631 | No clear effect |
| apply L_edit to more up blocks (Upblock1+2+3) (ablation) | MasterWeaver · CelebA-HQ single reference | Face similarity: 0.631 -> 0.568 | Hurt |
| increase lambda in identity cross attention (ablation) | MasterWeaver · CelebA-HQ single reference | identity metrics vs CLIP-T | Mixed |
| use MasterWeaver instead of FastComposer/IP-Adapter | single-reference personalization · CelebA-HQ single reference | CLIP-T: 0.201 -> 0.232 (vs FastComposer) | Improved |
| add more reference images at inference by concatenating identity tokens | MasterWeaver · multi-reference | identity fidelity | Improved |
Later papers that compared with this method
1 later accepted paper reports MasterWeaver as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- DynamicID: Zero-Shot Multi-ID Image Personalization with Flexible Facial Editability (ICCV 2025 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most 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 1 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
- When StyleGAN Meets Stable Diffusion: a W+ Adapter for Personalized Image Generation (CVPR 2024 · similarity 0.91)
- DreamSalon: A Staged Diffusion Framework for Preserving Identity-Context in Editable Face Generation (CVPR 2024 · similarity 0.89)
- DreamIdentity: Enhanced Editability for Efficient Face-Identity Preserved Image Generation (AAAI 2024 · similarity 0.89)
- InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning (CVPR 2024 · similarity 0.88)
- PortraitBooth: A Versatile Portrait Model for Fast Identity-preserved Personalization (CVPR 2024 · similarity 0.88)
- Identity-Preserving Text-to-Video Generation by Frequency Decomposition (CVPR 2025 · similarity 0.87)
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.
