Research map / Label-efficient and robust learning / Recognition and long-tail
FreeEnricher: Enriching Face Landmarks without Additional Cost
Read the paperCheck 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 |
|---|---|---|---|
| Add FreeEnricher to ADNet during training | ADNet · Enriched 300W | NME_point: 3.21 -> 3.06 | Helped |
| Add FreeEnricher to ADNet during training | ADNet · Enriched 300W | NME_edge: 1.18 -> 0.98 | Helped |
| Add FreeEnricher to ADNet during training | ADNet · 300W | NME: 2.93 -> 2.87 | Helped |
| Add FreeEnricher to ADNet during training | ADNet · WFLW | NME: 4.14 -> 4.10 | Helped |
| Regress offsets with FreeEnricher instead of using random offsets (ablation) | FreeEnricher offset refinement · 300W Testset | ME: 4.03 -> 1.92 | Helped |
| Regress offsets with FreeEnricher instead of using random offsets (ablation) | FreeEnricher offset refinement · WFLW Testset | ME: 3.96 -> 1.89 | Helped |
| Use B-spline rather than line interpolation for initialization (ablation) | FreeEnricher without other operators · Enriched 300W | NME: 3.18 -> 3.11 | Helped |
| Add patch normalizing to line initialization (ablation) | FreeEnricher · Enriched 300W | NME: 3.18 -> 3.14 | Helped |
| Add index embedding to line initialization (ablation) | FreeEnricher · Enriched 300W | NME: 3.18 -> 3.16 | Helped |
| Add quality scoring to line initialization and patch normalizing (ablation) | FreeEnricher · Enriched 300W | NME: 3.14 -> 3.11 | Helped |
| Add index embedding to the configuration with line initialization, patch normalizing, and quality scoring (ablation) | FreeEnricher · Enriched 300W | NME: 3.11 -> 3.06 | Helped |
| Increase enriching density from 1 to 10 (ablation) | FreeEnricher · Enriched 300W | NME: 2.91 -> 3.14 | Hurt |
Closest papers by meaning
- Generalizable Face Landmarking Guided by Conditional Face Warping (CVPR 2024 · similarity 0.88)
- Brute-Force Facial Landmark Analysis With a 140,000-Way Classifier (AAAI 2018 · similarity 0.87)
- Sparse Local Patch Transformer for Robust Face Alignment and Landmarks Inherent Relation Learning (CVPR 2022 · similarity 0.87)
- Face Alignment Across Large Poses: A 3D Solution (CVPR 2016 · similarity 0.87)
- LUVLi Face Alignment: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihood (CVPR 2020 · similarity 0.86)
- FaceLift: Semi-supervised 3D Facial Landmark Localization (CVPR 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.