Research map / 3D vision / Human pose estimation
Geometry-Driven Self-Supervised Method for 3D Human Pose 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 |
|---|---|---|---|
| add transform re-projection loss on top of re-projection loss (ablation) | self-supervised lifting network · Human3.6M Protocol 1 | MPJPE (mm): 468.5 -> 59.0 | Helped |
| replace adversarial loss with transform re-projection loss (ablation) | self-supervised lifting network · Human3.6M | MPJPE and P-MPJPE (mm): 106.6 -> 59.0 (Protocol 1); 78.4 -> 45.7 (Protocol 2) | Helped |
| add data augmentation with 8 virtual camera views (ablation) | Trans Reproj · Human3.6M | MPJPE and P-MPJPE (mm): 59.0 -> 57.0 (Protocol 1); 45.7 -> 44.1 (Protocol 2) | Helped |
| use TemporalDilated instead of ResLinear backbone (ablation) | Trans Reproj training · Human3.6M Protocol 1 | MPJPE (mm): 59.7 -> 56.1 | Helped |
| use TemporalDilated instead of ResLinear backbone (ablation) | Trans Reproj training · Human3.6M Protocol 2 | P-MPJPE (mm): 45.0 -> 43.2 | Helped |
| add network pre-training with multi-view consistency loss (ablation) | two-branch network · Human3.6M | MPJPE | Helped |
| compare with Kocabas et al. triangulation pseudo labels | self-supervised method · Human3.6M Protocol 1 | MPJPE (mm): 76.6 -> 59.0 | Helped |
| compare with Chen et al. on MPI-INF-3DHP | self-supervised method · MPI-INF-3DHP | PCK: 71.1 -> 74.1 | Helped |
Closest papers by meaning
- Unsupervised 3D Pose Estimation With Geometric Self-Supervision (CVPR 2019 · similarity 0.91)
- Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesis (CVPR 2020 · similarity 0.91)
- SelfPose3d: Self-Supervised Multi-Person Multi-View 3d Pose Estimation (CVPR 2024 · similarity 0.90)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · similarity 0.90)
- PandaPose: 3D Human Pose Lifting from a Single Image via Propagating 2D Pose Prior to 3D Anchor Space (NeurIPS 2025 · similarity 0.89)
- Lifting by Image – Leveraging Image Cues for Accurate 3D Human Pose Estimation (AAAI 2024 · similarity 0.89)
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.