Research map / 3D vision / Human pose estimation
Hierarchical Kinematic Human Mesh Recovery
Georgios Georgakis, Ren Li, Srikrishna Karanam, Terrence Chen, Jana Košecká, Ziyan Wu
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 |
|---|---|---|---|
| model all previous joints in chain (joint hierarchy) instead of only preceding joint (ablation) | HKMR · Human3.6M P1 | MPJPE (mm): 77.10 -> 71.08 | Improved |
| model all previous joints in chain instead of only preceding joint (ablation) | HKMR · Human3.6M P2 | MPJPE (mm): 74.28 -> 67.74 | Improved |
| add backward pass to forward-only inner iteration (ablation) | HKMR · Human3.6M P1 | MPJPE (mm): 75.99 -> 71.08 | Improved |
| add backward pass to forward-only inner iteration (ablation) | HKMR · Human3.6M P2 | MPJPE (mm): 72.10 -> 67.74 | Improved |
| use VAE pose prior instead of discriminator (ablation) | HKMR · Human3.6M P1 | MPJPE (mm): 74.21 -> 71.08 | Improved |
| use VAE pose prior instead of discriminator (ablation) | HKMR · Human3.6M P2 | MPJPE (mm): 71.72 -> 67.74 | Improved |
| increase outer iterations from 1 to 3 (ablation) | HKMR · Human3.6M P1 | MPJPE (mm): 93.79 -> 71.08 | Improved |
| increase outer iterations from 3 to 4 (ablation) | HKMR · Human3.6M P1 | MPJPE (mm): 71.08 -> 73.16 | Hurt |
| increase inner iterations from 2 to 4 (ablation) | HKMR · Human3.6M P1 | MPJPE (mm): 83.34 -> 71.08 | Improved |
| increase inner iterations from 4 to 6 (ablation) | HKMR · Human3.6M P1 | MPJPE (mm): 71.08 -> 73.61 | Hurt |
| replace HMR regressor with HKMR regressor | HKMR · Human3.6M P1 | MPJPE (mm): 87.97 -> 71.08 | Improved |
| replace CMR regressor with HKMR regressor | HKMR · Human3.6M P1 | MPJPE (mm): 74.70 -> 71.08 | Improved |
Closest papers by meaning
- Recovering 3D Human Mesh From Monocular Images: A Survey (TPAMI 2023 · similarity 0.92)
- Probabilistic Modeling for Human Mesh Recovery (ICCV 2021 · similarity 0.90)
- Skeleton2Mesh: Kinematics Prior Injected Unsupervised Human Mesh Recovery (ICCV 2021 · similarity 0.90)
- SAM 3D Body: Robust Full-Body Human Mesh Recovery (CVPR 2026 · similarity 0.90)
- Reconstructing Humans with a Biomechanically Accurate Skeleton (CVPR 2025 · similarity 0.89)
- FactorizedHMR: A Hybrid Framework for Video Human Mesh Recovery (NeurIPS 2026 · 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.
