Research map / Image generation / Text-to-image diffusion
Real-time 3D-aware Portrait Editing from a Single Image
Qingyan Bai, Zifan Shi, Yinghao Xu, Hao Ouyang, Qiuyu Wang, Ceyuan Yang, Xuan Wang, Gordon Wetzstein, Yujun Shen, Qifeng Chen
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 |
|---|---|---|---|
| distill into a feedforward module instead of 2D editing followed by Live3D inversion | IP2P+Live3D · FFHQ 100 pairs | Inference time: 10s -> 0.04s | Improved |
| distill into a feedforward module instead of IP2P+Live3D | IP2P+Live3D · FFHQ 100 pairs | CLIP_r: 0.62 -> 0.73 | Improved |
| 3DPE versus Live3D+StyleCLIP on identity | 3DPE · FFHQ 100 pairs | ID_t: 0.63 vs 0.47 | Hurt |
| add 3D distillation loss L_3D from Live3D (ablation) | 3DPE · FFHQ qualitative | geometry quality | Improved |
| add 2D distillation loss L_2D from diffusion prior (ablation) | 3DPE · FFHQ qualitative | texture detail | Improved |
| increase pretraining styles from 3 to 20 (ablation) | customized prompt adaptation · golden statue adaptation | LPIPS after 5min: 0.5220 -> 0.4849 | Improved |
| increase pretraining styles from 3 to 20 (ablation) | customized prompt adaptation · golden statue adaptation | LPIPS after 0.1min: 0.7758 -> 0.5703 | Improved |
| increase tuning pairs from 2 to 50 (ablation) | customized prompt adaptation · golden statue adaptation | LPIPS after 5min: 0.5212 -> 0.4797 | Improved |
| increase tuning pairs from 10 to 50 (ablation) | customized prompt adaptation · golden statue adaptation | LPIPS after 1min: 0.5242 -> 0.5718 | Hurt |
| train only prompt encoder and decoder while freezing the rest of Live3D | 3DPE · training | convergence speed | Improved |
Closest papers by meaning
- InstructPix2NeRF: Instructed 3D Portrait Editing from a Single Image (ICLR 2024 · similarity 0.88)
- Reference-Based 3D-Aware Image Editing with Triplanes (CVPR 2025 · similarity 0.86)
- 3D-Aware Face Editing via Warping-Guided Latent Direction Learning (CVPR 2024 · similarity 0.86)
- SHAP-EDITOR: Instruction-Guided Latent 3D Editing in Seconds (CVPR 2024 · similarity 0.86)
- StrokeFaceNeRF: Stroke-based Facial Appearance Editing in Neural Radiance Field (CVPR 2024 · similarity 0.85)
- Fast Text-to-3D-Aware Face Generation and Manipulation via Direct Cross-modal Mapping and Geometric Regularization (ICML 2024 · similarity 0.85)
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.
