Research map / 3D vision / Human pose estimation
Learning Pose Grammar to Encode Human Body Configuration for 3D Pose Estimation
Hao-Shu Fang, Yuanlu Xu, Wenguan Wang, Xiaobai Liu, Song-Chun Zhu
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 the full pose grammar (kinematics, symmetry, and motor coordination) (ablation) | Ours · Human3.6M Protocol #3 | Error (mm): 75.1 -> 72.8 | Improved |
| Add kinematics grammar (ablation) | Ours · Human3.6M Protocol #3 | Error (mm): 75.1 mm→73.9 mm | Improved |
| Add symmetry grammar to kinematics grammar (ablation) | Ours · Human3.6M Protocol #3 | Error (mm): 73.9 mm→73.2 mm | Improved |
| Add motor coordination grammar to kinematics and symmetry grammar (ablation) | Ours · Human3.6M Protocol #3 | Error (mm): 73.2 -> 72.8 | Improved |
| Add extra training pairs from six virtual camera views (ablation) | Ours · Human3.6M Protocol #3 | Error (mm): 82.6 mm→76.7 mm | Improved |
| Use PSS-generated detections instead of ground-truth 2D poses for the extra pairs (ablation) | Ours · Human3.6M Protocol #3 | Error (mm): 76.7 -> 72.8 | Improved |
| Replace PSS with simple random-noise augmentation (ablation) | Ours · Human3.6M Protocol #3 | Error (mm): 72.8 -> 78.0 | Hurt |
| Add PSS (ablation) | Bruce et al. · Human3.6M Protocol #3 | Error (mm): 112.1 -> 96.3 | Improved |
| Add PSS (ablation) | Martinez et al. · Human3.6M Protocol #3 | Error (mm): 84.9 -> 76.0 | Improved |
| Use the proposed pose-grammar and PSS method instead of Martinez et al. | Human3.6M Protocol #3 · Human3.6M Protocol #3 | Average Euclidean Distance (mm): 84.9 vs 72.8 | Improved |
| Use the proposed method instead of Martinez et al. | Human3.6M Protocol #1 · Human3.6M Protocol #1 | Average Euclidean Distance (mm): 62.9 -> 60.4 | Improved |
Closest papers by meaning
- Monocular 3D Pose Estimation via Pose Grammar and Data Augmentation (TPAMI 2021 · similarity 0.94)
- Progressive Bi-C3D Pose Grammar for Human Pose Estimation (AAAI 2020 · similarity 0.91)
- 3D Human Pose Estimation From a Single Image via Distance Matrix Regression (CVPR 2017 · similarity 0.90)
- A Simple yet Effective Baseline for 3D Human Pose Estimation (ICCV 2017 · similarity 0.89)
- 3D Human Pose Estimation = 2D Pose Estimation + Matching (CVPR 2017 · similarity 0.89)
- Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose (ECCV 2020 · similarity 0.89)
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.
