Research map / 3D vision / Human pose estimation
3DSA:Multi-View 3D Human Pose Estimation With 3D Space Attention Mechanisms
Po Han Chen, Chia-Chi Tsai
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 3D space attention (10x10x3) to VoxelPose | VoxelPose · CMU Panoptic | AP25: 83.59 -> 94.2 | Improved |
| add 3D space attention (10x10x3) to Faster VoxelPose | Faster VoxelPose · CMU Panoptic | MPJPE: 18.26mm -> 14.55mm | Improved |
| add 3DSA to VoxelPose | VoxelPose · CMU Panoptic | AP50/AP100/AP150: AP100 99.76 -> 99.21 | Mixed |
| finer space subdivision from 3x3x3 to 10x10x3 (ablation) | VoxelPose + 3DSA · CMU Panoptic | AP25: 92.73 -> 94.2 | Improved |
| subdivide further from 10x10x3 to 20x20x9 (ablation) | VoxelPose + 3DSA · CMU Panoptic | AP25: 94.2 -> 94.44 | No clear effect |
| predict attention scores from heatmaps instead of images (ablation) | VoxelPose + 3DSA · CMU Panoptic | AP25: 94.2 -> 86.97 | Hurt |
| add 3DSA with Heatmap-based input (synthetic) on Shelf/Campus | VoxelPose · Shelf and Campus | PCP average: Campus 96.7 -> 96.7; Shelf 97 -> 97.5 | No clear effect |
| add 3DSA with fewer cameras (3 cameras) | Faster VoxelPose · CMU Panoptic (3 cameras) | AP25: 53.68 -> 73.06 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Faster VoxelPose — Faster VoxelPose: Real-Time 3D Human Pose Estimation by Orthographic Projection (ECCV 2022) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Attention Mechanism Exploits Temporal Contexts: Real-Time 3D Human Pose Reconstruction (CVPR 2020 · similarity 0.87)
- MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimation (CVPR 2025 · similarity 0.87)
- Enhanced 3D Human Pose Estimation from Videos by Using Attention-Based Neural Network with Dilated Convolutions (IJCV 2021 · similarity 0.86)
- Unified 2D-3D Discrete Priors for Noise-Robust and Calibration-Free Multiview 3D Human Pose Estimation (NeurIPS 2025 · similarity 0.86)
- FreeMan: Towards Benchmarking 3D Human Pose Estimation under Real-World Conditions (CVPR 2024 · similarity 0.85)
- SVTformer: Spatial-View-Temporal Transformer for Multi-View 3D Human Pose Estimation (AAAI 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.
