Research map / Learning theory and algorithms / Prediction and decision losses
Look at Boundary: A Boundary-Aware Face Alignment Algorithm
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 |
|---|---|---|---|
| use boundary maps rather than facial-parts Gaussian cues (ablation) | BL · 300-W Challenging Subset | mean error (%): 5.25 -> 3.28 | Helped |
| use boundary maps rather than face-parsing cues (ablation) | BL · 300-W Challenging Subset | mean error (%): 4.16 -> 3.28 | Helped |
| add boundary information fusion (ablation) | LAB · COFW-29 testset | mean error (%): 5.58 -> 3.92 | Helped |
| add boundary information fusion (ablation) | LAB · AFLW-Full | mean error (%): 1.85 -> 1.25 | Helped |
| add boundary information fusion (ablation) | LAB · AFLW-Frontal | mean error (%): 1.62 -> 1.14 | Helped |
| fuse boundary heatmaps at progressively more regression levels (ablation) | BL · 300-W Challenging Subset | mean error (%): 7.12 -> 6.56 -> 6.32 -> 6.19 -> 6.13 | Helped |
| use a boundary-aware hourglass fusion subnet rather than an hourglass subnet without boundary features (ablation) | BL · 300-W Challenging Set | mean error (%): 6.95 -> 6.13 | Helped |
| use an hourglass fusion subnet rather than consecutive convolutional layers (ablation) | boundary fusion · 300-W Challenging Set | mean error (%): 6.24 -> 6.13 | Helped |
| add message passing between boundary heatmaps (ablation) | HBL · 300-W Challenging Set | mean error (%): 6.13 -> 5.82 | Helped |
| add message passing between boundary heatmaps (ablation) | HBL · 300-W Challenging Set | heatmap estimation error (%): 0.85 -> 0.76 | Helped |
| add adversarial learning with the landmark-based boundary effectiveness discriminator (ablation) | HBL+MP · 300-W Challenging Set | mean error (%): 5.82 -> 5.59 | Helped |
| add adversarial learning with the landmark-based boundary effectiveness discriminator (ablation) | HBL+MP · 300-W Challenging Set | heatmap estimation error (%): 0.76 -> 0.63 | Helped |
Closest papers by meaning
- LUVLi Face Alignment: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihood (CVPR 2020 · similarity 0.89)
- Brute-Force Facial Landmark Analysis With a 140,000-Way Classifier (AAAI 2018 · similarity 0.89)
- AnchorFace: An Anchor-based Facial Landmark Detector Across Large Poses (AAAI 2021 · similarity 0.87)
- Generalizable Face Landmarking Guided by Conditional Face Warping (CVPR 2024 · similarity 0.86)
- FaceLift: Semi-supervised 3D Facial Landmark Localization (CVPR 2024 · similarity 0.86)
- POPoS: Improving Efficient and Robust Facial Landmark Detection with Parallel Optimal Position Search (AAAI 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.