Research map / 3D vision / Human pose estimation
Object-Occluded Human Shape and Pose Estimation From a Single Color Image
Tianshu Zhang, Buzhen Huang, Yangang Wang
Check 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 |
|---|---|---|---|
| use the two-branch UV inpainting method instead of SPIN | human shape and pose estimation · Synthetic Occlusion Human3.6M | PA-MPJPE: 64.9 -> 56.4 | Improved |
| use the two-branch UV inpainting method instead of SPIN | human shape and pose estimation · 3DPW (occluded sequences) | PA-MPJPE: 95.4 -> 72.2 | Improved |
| use the two-branch UV inpainting method instead of SPIN | human shape and pose estimation · 3DOH50K | PA-MPJPE: 67.5 -> 58.5 | Improved |
| use the two-branch UV inpainting method instead of SPIN | human shape and pose estimation · 3DOH50K | Surface Error: 73.6 -> 63.3 | Improved |
| use the two-branch UV inpainting method instead of SPIN | human shape and pose estimation · Human3.6M (Protocol#2) | PA-MPJPE: 41.1 -> 41.7 | Hurt |
| add a UV map inpainting branch with latent-space supervision (ablation) | human shape and pose estimation network · 3DOH50K testing dataset | PA-MPJPE: 67.3 -> 56.4 | Improved |
| use latent-space supervision rather than cascading UV inpainting with the color image encoder (ablation) | human shape and pose estimation network · 3DOH50K testing dataset | PA-MPJPE: 61.9 -> 56.4 | Improved |
| add saliency map estimation and use the saliency map as input (ablation) | two-branch human shape and pose estimation network · 3DOH50K testing dataset | PA-MPJPE: 57.9 -> 56.4 | Improved |
| add the proposed part loss for connection vertices in UV maps (ablation) | UV map inpainting network · 3DOH50K | mesh connection smoothness | Improved |
Closest papers by meaning
- Object-Occluded Human Shape and Pose Estimation With Probabilistic Latent Consistency (TPAMI 2022 · similarity 0.98)
- Learning to Estimate 3D Human Pose and Shape From a Single Color Image (CVPR 2018 · similarity 0.90)
- Occlusion-Aware Networks for 3D Human Pose Estimation in Video (ICCV 2019 · similarity 0.89)
- Explicit Occlusion Reasoning for Multi-Person 3D Human Pose Estimation (ECCV 2022 · similarity 0.89)
- $\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.86)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.86)
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.
