Research map / Detection and segmentation / CNN features
Robust Facial Landmark Detection via Occlusion-Adaptive Deep 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 |
|---|---|---|---|
| use ODN instead of PCD-CNN | facial landmark detection · 300W Common set | NRMSE (×10−2): 3.67 -> 3.56 | Helped |
| use ODN instead of PCD-CNN | facial landmark detection · 300W Fullset | NRMSE (×10−2): 4.44 -> 4.17 | Helped |
| use ODN instead of SAN | facial landmark detection · 300W Challenging set | NRMSE (×10−2): 6.60 -> 6.67 | Hurt |
| use ODN instead of DSRN | facial landmark detection · AFLW-Full | NRMSE (×10−2): 1.86 -> 1.63 | Helped |
| use ODN instead of DAC-OSR | facial landmark detection · AFLW-Frontal | NRMSE (×10−2): 1.81 -> 1.38 | Helped |
| add geometry-aware and distillation modules (ablation) | BRNet · 300W Challenging set | NRMSE (×10−2): 7.21 -> 7.04 | Helped |
| add distillation and low-rank learning modules (ablation) | BRNet · 300W Challenging set | NRMSE (×10−2): 7.21 -> 6.88 | Helped |
| add geometry-aware and low-rank learning modules (ablation) | BRNet · 300W Challenging set | NRMSE (×10−2): 7.21 -> 6.90 | Helped |
| add the distillation module to geometry-aware and low-rank learning modules (ablation) | BRNet+GM+LM · 300W Challenging set | NRMSE (×10−2): 6.90 -> 6.67 | Helped |
| add the geometry-aware module to distillation and low-rank learning modules (ablation) | BRNet+DM+LM · 300W Challenging set | NRMSE (×10−2): 6.88 -> 6.67 | Helped |
| add the low-rank learning module to geometry-aware and distillation modules (ablation) | BRNet+GM+DM · 300W Challenging set | NRMSE (×10−2): 7.04 -> 6.67 | Helped |
| add L1 regularization to the distillation module (ablation) | BRNet+GM+LM+DM · 300W Challenging set | NRMSE (×10−2): 6.81 -> 6.67 | Helped |
Closest papers by meaning
- Occlusion-Free Face Alignment: Deep Regression Networks Coupled With De-Corrupt AutoEncoders (CVPR 2016 · similarity 0.89)
- Occlusion-Robust Face Alignment Using a Viewpoint-Invariant Hierarchical Network Architecture (CVPR 2022 · similarity 0.88)
- Deep Structured Prediction for Facial Landmark Detection (NeurIPS 2019 · similarity 0.88)
- Robust Facial Landmark Detection via a Fully-Convolutional Local-Global Context Network (CVPR 2018 · similarity 0.88)
- Simultaneous Facial Landmark Detection, Pose and Deformation Estimation Under Facial Occlusion (CVPR 2017 · similarity 0.87)
- OFER: Occluded Face Expression Reconstruction (CVPR 2025 · similarity 0.84)
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.