Research map / 3D vision / Human pose estimation
V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation From a Single Depth Map
Read the paperCheck 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 |
|---|---|---|---|
| Replace direct coordinate regression with per-voxel likelihood estimation using 2D depth input (ablation) | Pose estimation network · NYU Hand Pose Dataset | average 3D distance error (mm): 18.85 -> 13.01 | Helped |
| Replace 2D depth map input with a 3D voxelized grid for coordinate regression (ablation) | Pose estimation network · NYU Hand Pose Dataset | average 3D distance error (mm): 18.85 -> 16.78 | Helped |
| Use voxelized input and per-voxel likelihood output instead of coordinate regression (ablation) | V2V-PoseNet · NYU Hand Pose Dataset | average 3D distance error (mm): 16.78 -> 10.37 | Helped |
| Add localization refinement (ablation) | V2V-PoseNet · NYU Hand Pose Dataset | average 3D distance error (mm): 11.14 -> 9.22 | Helped |
| Add epoch ensemble (ablation) | V2V-PoseNet with localization refinement · NYU Hand Pose Dataset | average 3D distance error (mm): 9.22 -> 8.42 | Helped |
| Use V2V-PoseNet instead of Pose-REN | 3D hand pose estimation · ICVL Hand Posture Dataset | mean error (mm): 6.79 -> 6.28 | Helped |
| Use V2V-PoseNet instead of Pose-REN | 3D hand pose estimation · NYU Hand Pose Dataset | mean error (mm): 11.81 -> 8.42 | Helped |
| Use V2V-PoseNet instead of Pose-REN | 3D hand pose estimation · MSRA Hand Pose Dataset | mean error (mm): 8.65 -> 7.49 | Helped |
| Use V2V-PoseNet instead of REN-9x6x6 | 3D human pose estimation · ITOP front-view | mAP (Mean): 84.9 -> 88.74 | Helped |
| Use V2V-PoseNet instead of REN-9x6x6 | 3D human pose estimation · ITOP top-view | mAP (Mean): 75.5 -> 83.44 | Helped |
Closest papers by meaning
- HandVoxNet: Deep Voxel-Based Network for 3D Hand Shape and Pose Estimation From a Single Depth Map (CVPR 2020 · similarity 0.94)
- Augmented Skeleton Space Transfer for Depth-Based Hand Pose Estimation (CVPR 2018 · similarity 0.90)
- Crossing Nets: Combining GANs and VAEs With a Shared Latent Space for Hand Pose Estimation (CVPR 2017 · similarity 0.90)
- DeepHand: Robust Hand Pose Estimation by Completing a Matrix Imputed With Deep Features (CVPR 2016 · similarity 0.90)
- Reconstructing Hands in 3D with Transformers (CVPR 2024 · similarity 0.88)
- DVMNet: Computing Relative Pose for Unseen Objects Beyond Hypotheses (CVPR 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.