Research map / 3D vision / Face and shape
Towards Fast, Accurate and Stable 3D Dense Face Alignment
Jianzhu Guo, Xiangyu Zhu, Yang Yang, Fan Yang, 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 meta-joint optimization of fWPDC and VDC (ablation) | 3DDFA-V2 (MobileNet) · AFLW2000-3D | NME (%): 4.04 -> 3.73 | Improved |
| add meta-joint optimization of fWPDC and VDC (ablation) | 3DDFA-V2 (MobileNet) · AFLW | NME (%): 5.10 -> 4.64 | Improved |
| use meta-joint instead of vanilla-joint (best fixed beta=0.5) (ablation) | 3DDFA-V2 · AFLW2000-3D | NME (%): 3.80 -> 3.73 | Improved |
| add landmark-regression regularization to meta-joint (ablation) | 3DDFA-V2 · AFLW2000-3D | NME (%): 3.73 -> 3.59 | Improved |
| add landmark-regression regularization to fWPDC (ablation) | fWPDC · AFLW2000-3D | NME (%): 4.04 -> 3.89 | Improved |
| add landmark-regression regularization to fWPDC (ablation) | fWPDC · AFLW | NME (%): 5.10 -> 4.84 | Improved |
| add landmark-regression regularization to VDC (ablation) | VDC · AFLW2000-3D | NME (%): 5.23 -> 3.92 | Improved |
| use lrr. extra branch instead of landmark constraint on parameter regression (ablation) | Meta-joint · AFLW2000-3D | NME (%): 3.71 -> 3.59 | Improved |
| use lrr. instead of landmark constraint (ablation) | Meta-joint · AFLW | NME (%): 4.80 -> 4.50 | Improved |
| add short-video-synthesis to meta-joint+lrr. (ablation) | 3DDFA-V2 · Menpo-3D | NME (%): 1.86 -> 1.71 | Improved |
| add short-video-synthesis to meta-joint+lrr. (ablation) | 3DDFA-V2 · Menpo-3D | stability (%): 0.52 -> 0.48 | Improved |
| use short-video-synthesis instead of random rotations per mini-batch (ablation) | 3DDFA-V2 · Menpo-3D | NME / stability (%): 1.76 / 0.50 -> 1.71 / 0.48 | Improved |
Closest papers by meaning
- DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face Alignment (CVPR 2023 · similarity 0.88)
- 3D Face Tracking from 2D Video through Iterative Dense UV to Image Flow (CVPR 2024 · similarity 0.87)
- Face Alignment Across Large Poses: A 3D Solution (CVPR 2016 · similarity 0.87)
- Face Alignment in Full Pose Range: A 3D Total Solution (TPAMI 2017 · similarity 0.86)
- MMFace: A Multi-Metric Regression Network for Unconstrained Face Reconstruction (CVPR 2019 · similarity 0.86)
- Learning Dense Correspondence for NeRF-Based Face Reenactment (AAAI 2024 · 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.
