Research map / 3D vision / Human pose estimation
HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation
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 |
|---|---|---|---|
| Replace naive target vectors with adaptively updated vectors using reconstructed parent joints (ablation) | HybrIK · 3DPW test with joints jittered by ±10 mm | MPJPE (24 joints): 16.2 -> 9.8 | Helped |
| Replace naive target vectors with adaptively updated vectors using reconstructed parent joints (ablation) | HybrIK · 3DPW test with joints jittered by ±20 mm | MPJPE (24 joints): 34.0 -> 20.2 | Helped |
| Replace naive target vectors with adaptively updated vectors using reconstructed parent joints (ablation) | HybrIK · 3DPW test with joints jittered by ±30 mm | MPJPE (24 joints): 53.4 -> 31.2 | Helped |
| Predict twist angles rather than use random twist values (ablation) | HybrIK with ground-truth joints and shape · Reconstruction experiment | 14 LSP joints error: 40.0 -> 6.1 | Helped |
| Predict twist angles rather than use random twist values (ablation) | HybrIK with ground-truth joints and shape · Reconstruction experiment | Vertex error: 67.3 -> 10.0 | Helped |
| Use zero twist angles instead of estimated twist angles (ablation) | HybrIK with ground-truth joints and shape · Reconstruction experiment | Vertex error: 10.0 -> 12.1 | Hurt |
| Use zero shape parameters instead of estimated shape parameters (ablation) | HybrIK · 3DPW | MPJPE: 80.0 -> 81.1 | Hurt |
| Use zero shape parameters instead of estimated shape parameters (ablation) | HybrIK · 3DPW | PVE: 94.5 -> 95.4 | Hurt |
| Use ground-truth shape parameters instead of estimated shape parameters (ablation) | HybrIK · 3DPW | MPJPE: 80.0 -> 72.7 | Helped |
| Use ground-truth shape parameters instead of estimated shape parameters (ablation) | HybrIK · 3DPW | PVE: 94.5 -> 87.4 | Helped |
| Apply HybrIK to correct the predicted 3D pose (ablation) | 3D keypoint estimator · 3DPW | MPJPE (24 joints): 88.2 mm -> 79.2 mm | Helped |
| Apply SMPLify instead of HybrIK to correct the predicted pose (ablation) | 3D keypoint estimator · 3DPW | MPJPE (24 joints): 88.2 mm -> 114.3 mm | Hurt |
Closest papers by meaning
- NIKI: Neural Inverse Kinematics With Invertible Neural Networks for 3D Human Pose and Shape Estimation (CVPR 2023 · similarity 0.89)
- End-to-End Recovery of Human Shape and Pose (CVPR 2018 · similarity 0.89)
- PLIKS: A Pseudo-Linear Inverse Kinematic Solver for 3D Human Body Estimation (CVPR 2023 · similarity 0.89)
- SAM 3D Body: Robust Full-Body Human Mesh Recovery (CVPR 2026 · similarity 0.88)
- Reconstructing Humans with a Biomechanically Accurate Skeleton (CVPR 2025 · similarity 0.87)
- HiPoser: 3D Human Pose Estimation with Hierarchical Shared Learning at Parts-Level Using Inertial Measurement Units (AAAI 2025 · similarity 0.87)
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.