Research map / 3D vision / Face and shape
Learning High-Fidelity Face Texture Completion Without Complete Face Texture
Jongyoo Kim, Jiaolong Yang, Xin Tong
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 image-space discriminator (ablation) | DSD-GAN (texture-space discriminator only) · FaceScape | SSIM (avg): 0.6888 -> 0.7684 | Improved |
| add image-space discriminator (ablation) | DSD-GAN (texture-space discriminator only) · FaceScape | FID: 43.0560 -> 3.3124 | Improved |
| add image-space discriminator (ablation) | DSD-GAN (texture-space discriminator only) · FaceScape | Perceptual similarity (avg): 0.6663 -> 0.7767 | Improved |
| add texture-space discriminator (ablation) | DSD-GAN (image-space discriminator only) · FaceScape | SSIM (avg): 0.7582 -> 0.7684 | Improved |
| add texture-space discriminator (ablation) | DSD-GAN (image-space discriminator only) · FaceScape | FID: 9.9678 -> 3.3124 | Improved |
| add texture-space discriminator (ablation) | DSD-GAN (image-space discriminator only) · FaceScape | Perceptual similarity (avg): 0.7682 -> 0.7767 | Improved |
| add coordinate regression to texture-space discriminator (ablation) | DSD-GAN (texture-space discriminator only) · FaceScape | SSIM (avg): 0.6758 -> 0.6888 | Improved |
| add coordinate regression to texture-space discriminator (ablation) | DSD-GAN (texture-space discriminator only) · FaceScape | FID: 56.4223 -> 43.0560 | Improved |
| add coordinate regression to texture-space discriminator (ablation) | DSD-GAN (texture-space discriminator only) · FaceScape | Perceptual similarity (avg): 0.6290 -> 0.6663 | Improved |
| train with unpaired face photos (DSD-GAN) instead of supervised on complete BFM textures | texture completion generator · FaceScape | SSIM (avg): 0.7434 -> 0.7684 | Improved |
| train with unpaired face photos (DSD-GAN) instead of supervised on complete BFM textures | texture completion generator · FaceScape | FID: 17.2121 -> 3.3124 | Improved |
| train with unpaired face photos (DSD-GAN) instead of supervised on Multi-PIE textures | texture completion generator · FaceScape | FID: 30.3546 -> 3.3124 | Improved |
Closest papers by meaning
- OSTeC: One-Shot Texture Completion (CVPR 2021 · similarity 0.90)
- Generative Face Completion (CVPR 2017 · similarity 0.89)
- GANFIT: Generative Adversarial Network Fitting for High Fidelity 3D Face Reconstruction (CVPR 2019 · similarity 0.89)
- UV-GAN: Adversarial Facial UV Map Completion for Pose-Invariant Face Recognition (CVPR 2018 · similarity 0.88)
- Fast-GANFIT: Generative Adversarial Network for High Fidelity 3D Face Reconstruction (TPAMI 2021 · similarity 0.88)
- FreeUV: Ground-Truth-Free Realistic Facial UV Texture Recovery via Cross-Assembly Inference Strategy (CVPR 2025 · 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.
