Research map / 3D vision / Face and shape
Face Normals "In-The-Wild" Using Fully Convolutional Networks
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 |
|---|---|---|---|
| Train a fully convolutional face-normal network on facial data | facial normal estimation · Photoface | mean angular error: 24.2 ± 5.4 -> 22.0 ± 6.3 | Helped |
| Use cosine loss instead of smooth ℓ1 loss (ablation) | proposed ResNet architecture · Photoface | mean angular error: 22.0 ± 6.3 -> 21.5 ± 6.9 | Helped |
| Use cosine loss instead of smooth ℓ1 loss (ablation) | proposed ResNet architecture · Photoface | angular error < 20°: 36.63% -> 29.9% | Hurt |
| Replace ResNet with PixelNet (ablation) | normal estimation network trained with cosine loss · Photoface | mean angular error: 21.5 ± 6.9 -> 23.5 ± 6.3 | Hurt |
| Replace ResNet with PixelNet (ablation) | normal estimation network trained with cosine loss · Photoface | angular error < 20°: 29.9% -> 35.17% | Helped |
Closest papers by meaning
- Unsupervised Face Normalization With Extreme Pose and Expression in the Wild (CVPR 2019 · similarity 0.89)
- Learning Detailed Face Reconstruction From a Single Image (CVPR 2017 · similarity 0.88)
- Cross-Modal Deep Face Normals With Deactivable Skip Connections (CVPR 2020 · similarity 0.88)
- Photorealistic Facial Texture Inference Using Deep Neural Networks (CVPR 2017 · similarity 0.87)
- InverseFaceNet: Deep Monocular Inverse Face Rendering (CVPR 2018 · similarity 0.87)
- Learning Person-Specific Animatable Face Models from In-the-Wild Images via a Shared Base Model (CVPR 2025 · similarity 0.85)
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.