Research map / Image generation / Editing and style transfer
Generative Face Completion
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 local adversarial loss (ablation) | Face completion model · CelebA test, masks O1-O6 | SSIM: 0.798 -> 0.753 | Mixed |
| Add global adversarial loss after the local adversarial loss (ablation) | Face completion model · CelebA test, masks O1-O6 | identity distance: 0.775 -> 0.694 | Helped |
| Add semantic parsing loss (ablation) | Face completion model · CelebA test, masks O1-O6 | identity distance: 0.694 -> 0.602 | Helped |
| Add Poisson blending after M4 (ablation) | M4 face completion · CelebA test, masks O1-O6 | SSIM: 0.804 -> 0.824 | Helped |
| Use reconstruction loss only instead of adversarial and parsing losses (ablation) | Face completion model · CelebA test, masks O1-O6 | identity distance: 0.602 -> 0.763 | Hurt |
| Use reconstruction loss only (ablation) | Face completion model · CelebA test, masks O1-O6 | PSNR: 19.4 vs 18.9 | Mixed |
| Replace M4 with retrained ContextEncoder | Face completion model · CelebA test, masks O1-O6 | SSIM: 0.804 vs 0.772 | Mixed |
| Increase square mask size | M5 face completion model · CelebA test | completion performance | Hurt |
| Vary the noise filling the masked region | Face completion model · CelebA test | completion appearance diversity | Helped |
Closest papers by meaning
- Pluralistic Image Completion (CVPR 2019 · similarity 0.88)
- Large Scale Image Completion via Co-Modulated Generative Adversarial Networks (ICLR 2021 · similarity 0.88)
- FaceCom: Towards High-fidelity 3D Facial Shape Completion via Optimization and Inpainting Guidance (CVPR 2024 · similarity 0.86)
- Large Occluded Human Image Completion via Image-Prior Cooperating (AAAI 2024 · similarity 0.85)
- CapHuman: Capture Your Moments in Parallel Universes (CVPR 2024 · similarity 0.85)
- MaGIC: Multi-modality Guided Image Completion (ICLR 2024 · similarity 0.84)
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.