Research map / 3D vision / Face and shape
Rotate-and-Render: Unsupervised Photorealistic Face Rotation From Single-View Images
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 |
|---|---|---|---|
| Augment Res18 training data with rotated faces | Res18 trained on CASIA · LFW | accuracy (%): 98.77 -> 98.95 | Helped |
| Augment Res18 training data with rotated faces | Res18 trained on MS1MV2 · LFW | accuracy (%): 99.63 -> 99.68 | Helped |
| Augment Res18 training data with rotated faces | Res18 trained on CASIA · IJB-A | verification rate @FAR=.01: 90.57±1.2 -> 91.98±0.7 | Helped |
| Augment Res18 training data with rotated faces | Res18 trained on MS1MV2 · IJB-A | verification rate @FAR=.01: 97.28±0.6 -> 97.30±0.6 | No clear effect |
| Augment R18 training data with rotated faces | R18 trained on MS1MV2 · MegaFace 1-million | rank-1 identification accuracy (%): 97.00 -> 97.48 | Helped |
| Augment R50 training data with rotated faces | R50 trained on MS1MV2 · MegaFace 1-million | rank-1 identification accuracy (%): 98.26 -> 98.44 | Helped |
| Add VGG loss (ablation) | Rotate-and-Render · LFW | FID: 93.1 -> 83.1 | Helped |
| Add VGG loss (ablation) | Rotate-and-Render · IJB-A | FID: 126.5 -> 70.9 | Helped |
| Use multi-scale discriminator and feature matching loss instead of Pix2Pix PatchGAN with image-level L1 loss (ablation) | Rotate-and-Render · LFW | FID: 83.9 -> 83.1 | Helped |
| Use multi-scale discriminator and feature matching loss instead of Pix2Pix PatchGAN with image-level L1 loss (ablation) | Rotate-and-Render · IJB-A | FID: 132.3 -> 70.9 | Helped |
Closest papers by meaning
- Pose-Guided Photorealistic Face Rotation (CVPR 2018 · similarity 0.90)
- Unsupervised Depth Estimation, 3D Face Rotation and Replacement (NeurIPS 2018 · similarity 0.87)
- Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars (CVPR 2023 · similarity 0.87)
- FaceLit: Neural 3D Relightable Faces (CVPR 2023 · similarity 0.87)
- A 3D GAN for Improved Large-Pose Facial Recognition (CVPR 2021 · similarity 0.87)
- ControlFace: Harnessing Facial Parametric Control for Face Rigging (CVPR 2025 · similarity 0.86)
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.