Research map / 3D vision / Human pose estimation
Deep Kinematics Analysis for Monocular 3D Human Pose Estimation
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 the full kinematics pipeline instead of Pavllo et al.'s temporal baseline | Monocular 3D pose estimation · Human3.6M, Protocol 1, 9 frames | MPJPE (mm): 49.8 -> 45.6 | Helped |
| Add projection-based 2D refinement (ablation) | Pavllo et al. single-frame baseline · Human3.6M | MPJPE (mm): 51.8 -> 49.9 | Helped |
| Add 3D pose decomposition into length and direction estimation (ablation) | 2D-refined pose estimator · Human3.6M | MPJPE (mm): 49.9 -> 47.1 | Helped |
| Add 3D trajectory completion (ablation) | 2D-refined, decomposed pose estimator · Human3.6M | MPJPE (mm): 47.1 -> 45.6 | Helped |
| Add projection-based 2D refinement (ablation) | Pavllo et al. single-frame baseline · Human3.6M | PA-MPJPE (mm): 40.0 -> 39.4 | Helped |
| Add 3D pose decomposition into length and direction estimation (ablation) | 2D-refined pose estimator · Human3.6M | PA-MPJPE (mm): 39.4 -> 37.5 | Helped |
| Add 3D trajectory completion (ablation) | 2D-refined, decomposed pose estimator · Human3.6M | PA-MPJPE (mm): 37.5 -> 36.2 | Helped |
| Use 9 input frames instead of 1 (ablation) | Proposed estimator · Human3.6M, Protocol 1 | MPJPE (mm): 49.2 -> 45.6 | Helped |
| Use 9 input frames instead of 1 (ablation) | Proposed estimator · Human3.6M, Protocol 2 | PA-MPJPE (mm): 38.9 -> 36.2 | Helped |
| Use 9 input frames instead of 1 (ablation) | Proposed estimator · Human3.6M | MPJVE: 7.80 -> 2.01 | Helped |
| Add trajectory completion (ablation) | Proposed estimator · Human3.6M, 9 frames | MPJVE: 3.29 -> 2.01 | Helped |
| Use projection correspondence when refining 2D joints (ablation) | 2D pose correction module · Human3.6M | MPJPE (mm): 51.8 -> 49.9 | Helped |
Closest papers by meaning
- Sparseness Meets Deepness: 3D Human Pose Estimation From Monocular Video (CVPR 2016 · similarity 0.90)
- Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation (AAAI 2020 · similarity 0.90)
- Capturing the Motion of Every Joint: 3D Human Pose and Shape Estimation with Independent Tokens (ICLR 2023 · similarity 0.89)
- PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimation (CVPR 2023 · similarity 0.89)
- ManiPose: Manifold-Constrained Multi-Hypothesis 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.88)
- Normalizing Flows on the Product Space of SO(3) Manifolds for Probabilistic Human Pose Modeling (CVPR 2024 · similarity 0.87)
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.