Research map / 3D vision / Human pose estimation
Point-To-Pose Voting Based Hand Pose Estimation Using Residual Permutation Equivariant Layer
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 view normalization (ablation) | ResidualPEL detection · Hands2017Challenge | mean joint error (mm): 11.67 -> 9.93 | Helped |
| replace distribution detection with direct weighted regression (ablation) | ResidualPEL point-to-pose voting · Hands2017Challenge | mean joint error (mm): 9.93 -> 9.82 | Helped |
| replace distribution detection with direct weighted regression (ablation) | ResidualPEL point-to-pose voting · NYU, single view | mean joint error (mm): 9.25 -> 8.99 | Helped |
| increase input point count from 256 to 1024 (ablation) | ResidualPEL, Hands2017Challenge detection · Hands2017Challenge | mean joint error (mm): 11.34 -> 9.93 | Helped |
| increase input point count from 1024 to 2048 (ablation) | ResidualPEL, Hands2017Challenge detection · Hands2017Challenge | mean joint error (mm): 9.93 -> 9.93 | No clear effect |
| increase input point count from 256 to 1024 (ablation) | ResidualPEL, Hands2017Challenge regression · Hands2017Challenge | mean joint error (mm): 11.21 -> 9.82 | Helped |
| increase input point count from 1024 to 2048 (ablation) | ResidualPEL, Hands2017Challenge regression · Hands2017Challenge | mean joint error (mm): 9.82 -> 9.87 | Hurt |
| increase input point count from 256 to 1024 (ablation) | ResidualPEL, NYU detection · NYU, single view | mean joint error (mm): 9.82 -> 9.25 | Helped |
| increase input point count from 1024 to 2048 (ablation) | ResidualPEL, NYU detection · NYU, single view | mean joint error (mm): 9.25 -> 9.32 | Hurt |
| increase input point count from 256 to 1024 (ablation) | ResidualPEL, NYU regression · NYU, single view | mean joint error (mm): 9.45 -> 8.99 | Helped |
| increase input point count from 1024 to 2048 (ablation) | ResidualPEL, NYU regression · NYU, single view | mean joint error (mm): 8.99 -> 9.08 | Hurt |
| increase input point count from 256 to 1024 (ablation) | ResidualPEL, NYU regression · NYU, three-view training | mean joint error (mm): 9.05 -> 8.35 | Helped |
Closest papers by meaning
- Two Heads Are Better than One: Image-Point Cloud Network for Depth-Based 3D Hand Pose Estimation (AAAI 2023 · similarity 0.92)
- Hand PointNet: 3D Hand Pose Estimation Using Point Sets (CVPR 2018 · similarity 0.91)
- Neural Voting Field for Camera-Space 3D Hand Pose Estimation (CVPR 2023 · similarity 0.89)
- Dense 3D Regression for Hand Pose Estimation (CVPR 2018 · similarity 0.89)
- HandVoxNet: Deep Voxel-Based Network for 3D Hand Shape and Pose Estimation From a Single Depth Map (CVPR 2020 · similarity 0.89)
- HandDiff: 3D Hand Pose Estimation with Diffusion on Image-Point Cloud (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.