Research map / 3D vision / Human pose estimation
PLIKS: A Pseudo-Linear Inverse Kinematic Solver for 3D Human Body Estimation
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 |
|---|---|---|---|
| Use PLIKS instead of the strongest listed baseline | 3DPW · 3DPW (14) | PA-MPJPE: 43.0 -> 38.5 | Helped |
| Use PLIKS instead of the strongest listed baseline | MPI-INF-3DHP · MPI-INF-3DHP (17) | MPJPE: 93.9 -> 67.6 | Helped |
| Use PLIKS instead of the strongest listed baseline | AGORA test · AGORA test | MPJPE: 81.0 -> 71.5 | Helped |
| Add training with 3DPW and AGORA (ablation) | PLIKS · 3DPW (14) | PA-MPJPE: 42.8 -> 40.1 | Helped |
| Use the HRNet-W48 backbone instead of the standard backbone (ablation) | PLIKS trained with 3DPW and AGORA · 3DPW (14) | PA-MPJPE: 40.1 -> 38.5 | Helped |
| Add CamCalib focal-length estimation at evaluation (ablation) | PLIKS · AGORA test | NMJE: 83.0 -> 79.6 | Helped |
| Run the PLIKS module twice to refine rotations (ablation) | PLIKS with GT correspondences · AGORA | MPJPE: 8.4 -> 1.1 | Helped |
| Replace HRNet-W32 with ResNet-50 (ablation) | ARE · 3DPW | MPJPE: 83.6 -> 87.1 | Hurt |
| Use PLIKS instead of ARE alone (ablation) | PLIKS · Human3.6M | MPJPE: 53.6 -> 51.3 | Helped |
| Use estimated focal length f=√(W²+H²) instead of weak perspective (ablation) | PLIKS · MPI-INF-3DHP | MPJPE: 93.4 -> 83.2 | Helped |
| Use estimated rather than known camera intrinsics (ablation) | PLIKS · MPI-INF-3DHP | MPJPE: 76.9 -> 83.2 | Hurt |
| Use estimated rather than known camera intrinsics (ablation) | PLIKS · 3DPW | MPJPE: 81.8 -> 81.8 | No clear effect |
Closest papers by meaning
- Reconstructing Humans with a Biomechanically Accurate Skeleton (CVPR 2025 · similarity 0.87)
- PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025 · similarity 0.85)
- MeshPose: Unifying DensePose and 3D Body Mesh Reconstruction (CVPR 2024 · similarity 0.85)
- RoGSplat: Learning Robust Generalizable Human Gaussian Splatting from Sparse Multi-View Images (CVPR 2025 · similarity 0.85)
- BLADE: Single-view Body Mesh Estimation through Accurate Depth Estimation (CVPR 2025 · similarity 0.85)
- VS: Reconstructing Clothed 3D Human from Single Image via Vertex Shift (CVPR 2024 · similarity 0.85)
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.