Research map / Image generation / Editing and style transfer
FaceComposer: A Unified Model for Versatile Facial Content Creation
Jiayu Wang, Kang Zhao, Yifeng Ma, Shiwei Zhang, Yingya Zhang, Yujun Shen, Deli Zhao, Jingren Zhou
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 |
|---|---|---|---|
| remove image data (train on videos only) (ablation) | FaceComposer · HDTF face animation | M-LMD: 2.25 -> 2.46 | Hurt |
| remove video data (train on images only) (ablation) | FaceComposer · HDTF face animation | F-LMD: 1.84 -> 1.94 | Hurt |
| remove video data (train on images only) (ablation) | FaceComposer · HDTF face animation | CPBD: 0.27 -> 0.29 | No clear effect |
| remove image data (ablation) | FaceComposer · HDTF face animation | SSIM: 0.78 -> 0.78 | No clear effect |
| drop ID condition (ablation) | FaceComposer · Face generation | FID: 11.34 -> 11.40 | No clear effect |
| drop ID and Sketch conditions (ablation) | FaceComposer · Face generation | FID: 11.34 -> 11.27 | No clear effect |
| drop ID and Sketch conditions (ablation) | FaceComposer · Face animation | F-LMD: 1.84 -> 1.88 | Hurt |
| reduce data to 10 h video and 4.5W images (ablation) | FaceComposer- · Face generation | FID: 11.34 -> 11.60 | Hurt |
| reduce data to 10 h video and 4.5W images (ablation) | FaceComposer- · Face animation | M-LMD: 2.25 -> 2.37 | Hurt |
| FaceComposer instead of LAFITE | face generation · Multi-Modal CelebA-HQ | R-precision (%): 81.94 -> 86.63 | Improved |
| FaceComposer instead of StyleClip | face editing · Edited images | IDS: 0.75 -> 0.94 | Improved |
| FaceComposer instead of Wav2Lip | face animation · HDTF | Syncconf: 5.27 -> 4.27 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TediGAN — TediGAN: Text-Guided Diverse Face Image Generation and Manipulation (CVPR 2021) · 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
- DreamSalon: A Staged Diffusion Framework for Preserving Identity-Context in Editable Face Generation (CVPR 2024 · similarity 0.88)
- Text-Guided 3D Face Synthesis - From Generation to Editing (CVPR 2024 · similarity 0.87)
- FaceChain-ImagineID: Freely Crafting High-Fidelity Diverse Talking Faces from Disentangled Audio (CVPR 2024 · similarity 0.87)
- Limitations of Face Image Generation (AAAI 2024 · similarity 0.87)
- Faces that Speak: Jointly Synthesising Talking Face and Speech from Text (CVPR 2024 · similarity 0.87)
- When StyleGAN Meets Stable Diffusion: a W+ Adapter for Personalized Image Generation (CVPR 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.
