Research map / 3D vision / Human pose estimation
Pose-Dependent Low-Rank Embedding for Head Pose Estimation
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 PLRE instead of SML-W | pose estimation · CMU-PIE, 0% noise | MAE: 5.547 -> 4.395 | Helped |
| Use PLRE instead of SML-W | pose estimation · CMU-PIE, 25% noise | MAE: 10.248 -> 7.363 | Helped |
| Use PLRE instead of SML-W | pose estimation · MIT-CBCL, 10% noise | MAE: 1.273 -> 0.024 | Helped |
| Use PLRE instead of SML-W | pose classification · CMU-PIE, 25% noise | accuracy: 80.91 -> 82.79 | Helped |
| Use PLRE instead of SML-W | pose classification · extended YaleB, 0% noise | accuracy: 86.34 -> 88.17 | Helped |
| Use PLRE instead of SML-W | pose classification · extended YaleB, 25% noise | accuracy: 83.42 -> 84.26 | Helped |
| Use supervised embeddings (SLEA, SLPP, SML-B, SML-W, PLRE) instead of unsupervised NPE/LPP | pose estimation · CMU-PIE, 25% noise | MAE: 12.816 -> 10.416 | Helped |
| Increase training label noise from 0% to 25% | PLRE · CMU-PIE | MAE: 4.395 -> 7.363 | Hurt |
| Increase training label noise from 0% to 20% | PLRE · CMU-PIE | accuracy: 91.94 -> 83.72 | Hurt |
| Increase training noise from 0% to 25% | SML-W · MIT-CBCL | MAE: 0.020 -> 2.003 | Hurt |
| Compare MAE vs accuracy ranking of SLEA and SLPP at 5% noise | baselines · CMU-PIE, 5% noise | MAE vs accuracy: 7.940 vs 7.584 MAE; 87.75 vs 84.93 ACC | Mixed |
Closest papers by meaning
- Unsupervised Discovery of Facial Landmarks and Head Pose (CVPR 2025 · similarity 0.86)
- Thin-Slicing for Pose: Learning to Understand Pose Without Explicit Pose Estimation (CVPR 2016 · similarity 0.86)
- TokenHPE: Learning Orientation Tokens for Efficient Head Pose Estimation via Transformers (CVPR 2023 · similarity 0.86)
- Dual Pose-invariant Embeddings: Learning Category and Object-specific Discriminative Representations for Recognition and Retrieval (CVPR 2024 · similarity 0.85)
- PoseBH: Prototypical Multi-Dataset Training Beyond Human Pose Estimation (CVPR 2025 · similarity 0.84)
- Can Generative Video Models Help Pose Estimation? (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.