Research map / 3D vision / Face and shape
Geometry-Aware Face Completion and Editing
Read the paperCheck 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 |
|---|---|---|---|
| use FCENet with estimated facial geometry guidance and mask disentanglement instead of GFC | Face completion · Multi-PIE testing set | PSNR: 27.208 -> 27.714 | Helped |
| use FCENet with estimated facial geometry guidance and mask disentanglement instead of GFC | Face completion · Multi-PIE testing set | SSIM: 0.889 -> 0.904 | Helped |
| use FCENet with estimated facial geometry guidance and mask disentanglement instead of GFC | Face completion · CelebA testing set | PSNR: 24.281 -> 24.944 | Helped |
| use FCENet with estimated facial geometry guidance and mask disentanglement instead of GFC | Face completion · CelebA testing set | SSIM: 0.836 -> 0.871 | Helped |
| add low-rank mask regularization (ablation) | FCENet without facial geometry inputs · Multi-PIE testing set | PSNR: 24.926 -> 25.470 | Helped |
| add low-rank mask regularization (ablation) | FCENet without facial geometry inputs · Multi-PIE testing set | SSIM: 0.841 -> 0.845 | Helped |
| add landmark heatmaps (ablation) | FCENet with low-rank loss and no geometry inputs · Multi-PIE testing set | PSNR: 25.470 -> 27.256 | Helped |
| add landmark heatmaps (ablation) | FCENet with low-rank loss and no geometry inputs · Multi-PIE testing set | SSIM: 0.845 -> 0.898 | Helped |
| add parsing maps (ablation) | FCENet with low-rank loss and no geometry inputs · Multi-PIE testing set | PSNR: 25.470 -> 27.482 | Helped |
| add parsing maps (ablation) | FCENet with low-rank loss and no geometry inputs · Multi-PIE testing set | SSIM: 0.845 -> 0.900 | Helped |
| add both landmark heatmaps and parsing maps (ablation) | FCENet with low-rank loss and no geometry inputs · Multi-PIE testing set | PSNR: 25.470 -> 27.714 | Helped |
| add both landmark heatmaps and parsing maps (ablation) | FCENet with low-rank loss and no geometry inputs · Multi-PIE testing set | SSIM: 0.845 -> 0.904 | Helped |
Closest papers by meaning
- FaceCom: Towards High-fidelity 3D Facial Shape Completion via Optimization and Inpainting Guidance (CVPR 2024 · similarity 0.87)
- Generative Face Completion (CVPR 2017 · similarity 0.86)
- FreeEnricher: Enriching Face Landmarks without Additional Cost (AAAI 2023 · similarity 0.85)
- 3D-Aware Face Editing via Warping-Guided Latent Direction Learning (CVPR 2024 · similarity 0.85)
- Generalizable Face Landmarking Guided by Conditional Face Warping (CVPR 2024 · similarity 0.84)
- StrokeFaceNeRF: Stroke-based Facial Appearance Editing in Neural Radiance Field (CVPR 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.