Research map / 3D vision / Face and shape
Beyond 3DMM Space: Towards Fine-grained 3D Face Reconstruction
Xiangyu Zhu, Fan Yang, Di Huang, Chang Yu, Hao Wang, Jianzhu Guo, Zhen Lei, Stan Z. Li
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 PNCC to the image input (ablation) | FGNet-CV · FG3D-test | NME (%): 3.37 -> 3.23 | Improved |
| add UV-visibility to UV-texture input (ablation) | FGNet-MV · FG3D-test | NME (%): 3.27 -> 3.19 | Improved |
| add UV-shape to UV-texture and UV-visibility (ablation) | FGNet-MV · FG3D-test | NME (%): 3.19 -> 3.10 | Improved |
| warp the image to UV space (model view) instead of camera view | FGNet · FG3D-test | NME (%): 3.23 -> 3.10 | Improved |
| apply FGNet-MV to initial 3DMM fit | 3DDFA initial shape · FG3D-test | NME (%): 5.07 -> 3.10 | Improved |
| apply FGNet-MV, error parallel to viewing direction | 3DDFA initial shape · FG3D-test | NME (%): 3.11 -> 1.59 | Improved |
| apply FGNet-MV, error orthogonal to viewing direction | 3DDFA initial shape · FG3D-test | NME (%): 3.44 -> 2.38 | Improved |
| use FGNet-MV versus Deng et al. cross-dataset | Deng et al. · Florence | NME (%): 6.12 -> 5.56 | Improved |
Closest papers by meaning
- Beyond 3DMM: Learning to Capture High-Fidelity 3D Face Shape (TPAMI 2022 · similarity 0.89)
- DF2Net: A Dense-Fine-Finer Network for Detailed 3D Face Reconstruction (ICCV 2019 · similarity 0.88)
- FaceVerse: A Fine-Grained and Detail-Controllable 3D Face Morphable Model From a Hybrid Dataset (CVPR 2022 · similarity 0.87)
- Uncertainty-Aware Mesh Decoder for High Fidelity 3D Face Reconstruction (CVPR 2020 · similarity 0.86)
- 3D Guided Fine-Grained Face Manipulation (CVPR 2019 · similarity 0.86)
- FaceScape: 3D Facial Dataset and Benchmark for Single-View 3D Face Reconstruction (TPAMI 2023 · 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.
