Research map / 3D vision / Human pose estimation
Sparseness Meets Deepness: 3D Human Pose Estimation From Monocular Video
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 |
|---|---|---|---|
| Add the sparse 3D pose prior and sequence optimization | 3D reconstruction from known 2D poses · Human3.6M, original 2D poses | mean per joint error (mm): 72.98 -> 49.64 | Helped |
| Add the sparse 3D pose prior and sequence optimization | 3D reconstruction from known 2D poses · Human3.6M, synthesized 2D poses | mean per joint error (mm): 48.88 -> 47.57 | Helped |
| Add EM inference using joint-location uncertainty and temporal smoothness | Proposed 3D pose estimation · Human3.6M, unknown 2D poses | mean per joint error (mm): 143.85 -> 125.55 | Helped |
| Add EM inference using joint-location uncertainty and temporal smoothness | Proposed 2D joint localization · Human3.6M, unknown 2D poses | 2D joint error (pixel): 15.00 -> 10.85 | Helped |
| Add 3D geometric and smoothness priors during joint localization | Proposed 2D pose estimation · PennAction golf swing | per joint distance error (pixels): 18.73 -> 14.03 | Helped |
| Add 3D geometric and smoothness priors during joint localization | Proposed 2D pose estimation · PennAction tennis forehand | per joint distance error (pixels): 25.75 -> 20.99 | Helped |
| Add temporal smoothness constraints (ablation) | Proposed 3D pose estimation · Human3.6M, unknown 2D poses | mean per joint error (mm): 120.99 -> 113.01 | Helped |
| Add temporal smoothness constraints (ablation) | Proposed 2D joint localization · Human3.6M, unknown 2D poses | 2D joint error (pixel): 11.25 -> 10.85 | Helped |
| Use action-specific CNN and pose dictionaries rather than one shared model (ablation) | Proposed 3D pose estimation · Human3.6M, unknown 2D poses | mean per joint error (mm): 116.49 -> 113.01 | Helped |
| Refine the rotation estimate with perspective adjustment | Proposed 3D pose estimation · Human3.6M, unknown 2D poses | mean per joint error (mm): 125.55 -> 113.01 | Helped |
Closest papers by meaning
- 3D Human Pose Estimation From a Single Image via Distance Matrix Regression (CVPR 2017 · similarity 0.91)
- Generating Multiple Hypotheses for 3D Human Pose Estimation With Mixture Density Network (CVPR 2019 · similarity 0.91)
- Deep Kinematics Analysis for Monocular 3D Human Pose Estimation (CVPR 2020 · similarity 0.90)
- Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation (AAAI 2020 · similarity 0.90)
- Uncertainty-Aware Adaptation for Self-Supervised 3D Human Pose Estimation (CVPR 2022 · similarity 0.90)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · similarity 0.88)
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.