Research map / 3D vision / Human pose estimation
Skeleton-Aware 3D Human Shape Reconstruction From Point Clouds
Haiyong Jiang, Jianfei Cai, Jianmin Zheng
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 graph aggregation (GA) module to PointNet++ (ablation) | Ours · Synthesized test set | mean distance (mm): 20.6 -> 15.5 | Improved |
| add graph aggregation (GA) module to PointNet++ (ablation) | Ours · Dyna jumping jacks | mean distance (mm): 31.3 -> 26.9 | Improved |
| add graph aggregation (GA) module to PointNet++ (ablation) | Ours · Dyna running on spot | mean distance (mm): 30.9 -> 22.5 | Improved |
| use attention module instead of MLP mapping (ablation) | Ours · Synthesized test set | mean distance (mm): 36.0 -> 15.5 | Improved |
| use attention module instead of MLP mapping (ablation) | Ours · Dyna running on spot | mean distance (mm): 42.7 -> 22.5 | Improved |
| use attention module instead of pooling and replication (ablation) | Ours · Synthesized test set | mean distance (mm): 141.1 -> 15.5 | Improved |
| use skeleton graph module instead of MLP (ablation) | Ours · Synthesized test set | mean distance (mm): 26.0 -> 15.5 | Improved |
| use skeleton graph module instead of MLP (ablation) | Ours · Dyna running on spot | mean distance (mm): 38.0 -> 22.5 | Improved |
| use AM and SGM instead of pooling plus MLP for SMPL parameter prediction (ablation) | Ours · Synthesized test set | mean distance (mm): 29.5 -> 15.5 | Improved |
| use AM and SGM instead of pooling plus MLP for SMPL parameter prediction (ablation) | Ours · Dyna jumping jacks | mean distance (mm): 41.0 -> 26.9 | Improved |
| add online fine tuning (ablation) | Ours · Dyna jumping jacks | mean distance (mm): 26.9 -> 14.2 | Improved |
| add online fine tuning (ablation) | Ours · Dyna running on spot | mean distance (mm): 22.5 -> 11.4 | Improved |
Closest papers by meaning
- Keep It SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image (ECCV 2016 · similarity 0.92)
- Convolutional Mesh Regression for Single-Image Human Shape Reconstruction (CVPR 2019 · similarity 0.90)
- Learning to Estimate 3D Human Pose and Shape From a Single Color Image (CVPR 2018 · similarity 0.90)
- Human as Points: Explicit Point-Based 3D Human Reconstruction From Single-View RGB Images (TPAMI 2025 · similarity 0.90)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 2024 · similarity 0.89)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 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.
