Research map / 3D vision / Human pose estimation
Motion Guided 3D Pose Estimation from Videos
Jingbo Wang, Sijie Yan, Yuanjun Xiong, Dahua Lin
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 cross-product motion loss (interval 12) (ablation) | UGCN · Human3.6M (GT 2D) | MPJPE: 32.0 -> 27.1 | Improved |
| add multi-scale motion loss intervals (8,12,16,24) instead of single scale (ablation) | UGCN · Human3.6M (GT 2D) | MPJPE: 27.1 -> 25.6 | Improved |
| use cross product instead of subtraction operator (ablation) | UGCN motion loss · Human3.6M (GT 2D) | MPJPE: 28.9 -> 27.1 | Improved |
| use cross product instead of inner product (ablation) | UGCN motion loss · Human3.6M (GT 2D) | MPJPE: 29.8 -> 27.1 | Improved |
| combine subtraction and inner product with cross product (ablation) | UGCN motion loss · Human3.6M (GT 2D) | MPJPE: 27.1 -> 27.1 | No clear effect |
| use motion loss instead of derivative loss (ablation) | UGCN · Human3.6M (GT 2D) | MPJPE: 31.6 -> 27.1 | Improved |
| add U-shaped down/upsampling (4 pairs) (ablation) | UGCN without motion loss · Human3.6M (GT 2D) | MPJPE: 38.6 -> 32.0 | Improved |
| add U-shaped down/upsampling (4 pairs) (ablation) | UGCN with motion loss · Human3.6M (GT 2D) | MPJPE: 36.9 -> 27.1 | Improved |
| add spatial graph (ablation) | UGCN · Human3.6M (GT 2D) | MPJPE: 39.2 -> 32.0 | Improved |
| add merging stage (ablation) | UGCN · Human3.6M (GT 2D) | MPJPE: 32.5 -> 32.0 | Improved |
| add merging stage (ablation) | UGCN with motion loss · Human3.6M (GT 2D) | MPJPE: 28.4 -> 25.6 | Improved |
| add motion loss | UGCN · MPI-INF-3DHP | MPJPE: 76.7 -> 68.1 | Improved |
Closest papers by meaning
- Learning 3D Human Pose from Structure and Motion (ECCV 2018 · similarity 0.90)
- Locally Connected Network for Monocular 3D Human Pose Estimation (TPAMI 2020 · similarity 0.89)
- Capturing Humans in Motion: Temporal-Attentive 3D Human Pose and Shape Estimation From Monocular Video (CVPR 2022 · similarity 0.88)
- Optimizing Network Structure for 3D Human Pose Estimation (ICCV 2019 · similarity 0.88)
- Graph and Temporal Convolutional Networks for 3D Multi-person Pose Estimation in Monocular Videos (AAAI 2021 · similarity 0.88)
- 3D Human Pose Estimation Using Spatio-Temporal Networks with Explicit Occlusion Training (AAAI 2020 · 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.
