Research map / 3D vision / Human pose estimation
Faster VoxelPose: Real-Time 3D Human Pose Estimation by Orthographic Projection
Hang Ye, Wentao Zhu, Chunyu Wang, Rujie Wu, Yizhou Wang
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 person-specific feature masking (ablation) | Faster VoxelPose · CMU Panoptic | MPJPE: 21.07mm -> 18.26mm | Improved |
| add person-specific feature masking (ablation) | Faster VoxelPose · CMU Panoptic | AP25: 72.05 -> 85.22 | Improved |
| add adaptive weighted fusion instead of mean (ablation) | Faster VoxelPose · CMU Panoptic | MPJPE: 20.11mm -> 18.26mm | Improved |
| add adaptive weighted fusion instead of mean (ablation) | Faster VoxelPose · CMU Panoptic | AP25: 77.23 -> 85.22 | Improved |
| increase cameras from 4 to 5 (ablation) | Faster VoxelPose · CMU Panoptic | MPJPE: 21.12mm -> 18.26mm | Improved |
| increase cameras from 3 to 5 (ablation) | Faster VoxelPose · CMU Panoptic | MPJPE: 26.13mm -> 18.26mm | Improved |
| use finer JLN voxels 64^3 instead of 48^3 (ablation) | Faster VoxelPose · CMU Panoptic | MPJPE: 19.66mm -> 18.26mm | Improved |
| use finer JLN voxels 64^3 instead of 32^3 (ablation) | Faster VoxelPose · CMU Panoptic | MACs: 2.167G -> 8.670G MACs; MPJPE 20.47mm -> 18.26mm | Mixed |
| replace 3D-CNN VoxelPose with orthographic projection and 2D-CNN | Faster VoxelPose · CMU Panoptic | inference time: 316.0ms -> 32.2ms | Improved |
| replace 3D-CNN VoxelPose with orthographic projection and 2D-CNN | Faster VoxelPose · CMU Panoptic | MPJPE: 17.68mm -> 18.26mm | Hurt |
| replace VoxelPose with Faster VoxelPose | Faster VoxelPose · Shelf | average PCP3D: 97.0 -> 97.6 | Improved |
| replace VoxelPose with Faster VoxelPose | Faster VoxelPose · Campus | average PCP3D: 96.7 -> 96.2 | Hurt |
Later papers that compared with this method
3 later accepted papers report Faster VoxelPose as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- 3DSA:Multi-View 3D Human Pose Estimation With 3D Space Attention Mechanisms (ECCV 2024 · better on 3 of 3 shared results)
- AvatarPose: Avatar-guided 3D Pose Estimation of Close Human Interaction from Sparse Multi-view Videos (ECCV 2024 · better on 1 of 1 shared result)
- TEMPO: Efficient Multi-View Pose Estimation, Tracking, and Forecasting (ICCV 2023 · better on 1 of 2 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
- VoxelTrack: Multi-Person 3D Human Pose Estimation and Tracking in the Wild (TPAMI 2022 · similarity 0.90)
- VoxelPose: Towards Multi-Camera 3D Human Pose Estimation in Wild Environment (ECCV 2020 · similarity 0.90)
- MetaPose: Fast 3D Pose From Multiple Views Without 3D Supervision (CVPR 2022 · similarity 0.88)
- A Simple yet Effective Baseline for 3D Human Pose Estimation (ICCV 2017 · similarity 0.88)
- FreeMan: Towards Benchmarking 3D Human Pose Estimation under Real-World Conditions (CVPR 2024 · similarity 0.87)
- 3D Human Pose Perception from Egocentric Stereo Videos (CVPR 2024 · similarity 0.86)
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.
