Research map / Image generation / Text-to-image diffusion
OmniPortrait: Fine-Grained Personalized Portrait Synthesis via Pivotal Optimization
Dongxu Yue, Bo Lin, Yao Tang, Jiajun Liang, Zhihai He, Chun Yuan
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 face localization loss to Pivot ID Encoder training (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | SIM: 35.01 -> 69.13 | Improved |
| add face localization loss to Pivot ID Encoder training (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | FID: 476.22 -> 213.48 | Improved |
| add Pivot ID Encoder (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | SIM: 14.15 -> 69.13 | Improved |
| add Pivot ID Encoder (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | CLIP-I: 21.83 -> 73.08 | Improved |
| add background gradient masking (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | SIM: 33.42 -> 69.13 | Improved |
| add background gradient masking (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | CLIP-T: 19.34 -> 24.25 | Improved |
| add RB-Guidance (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | SIM: 63.87 -> 69.13 | Improved |
| add RB-Guidance (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | CLIP-I: 66.10 -> 73.08 | Improved |
| add RB-Guidance (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | CLIP-T: 24.88 -> 24.25 | Hurt |
| add RB-Guidance (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | FID: 210.45 -> 213.48 | Hurt |
| use guidance start time u = 0.6 rather than earlier or later (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | identity similarity: u = 0.6 | Mixed |
| larger face localization loss weight eta (ablation) | OmniPortrait · Ablation (SDXL, 50 IDs) | SIM / CLIP-T: higher face similarity, lower text alignment | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- PhotoMaker — PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding (CVPR 2024) · better on 3 of 4 shared results
- FastComposer — FastComposer: Tuning-Free Multi-subject Image Generation with Localized Attention (IJCV 2024) · better on 4 of 4 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
- RealisID: Scale-Robust and Fine-Controllable Identity Customization via Local and Global Complementation (AAAI 2025 · similarity 0.88)
- PortraitBooth: A Versatile Portrait Model for Fast Identity-preserved Personalization (CVPR 2024 · similarity 0.88)
- Omni-ID: Holistic Identity Representation Designed for Generative Tasks (CVPR 2025 · similarity 0.88)
- When StyleGAN Meets Stable Diffusion: a W+ Adapter for Personalized Image Generation (CVPR 2024 · similarity 0.87)
- DreamSalon: A Staged Diffusion Framework for Preserving Identity-Context in Editable Face Generation (CVPR 2024 · similarity 0.87)
- MagiCapture: High-Resolution Multi-Concept Portrait Customization (AAAI 2024 · 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.
