Research map / 3D vision / Human pose estimation
Playing for 3D Human Recovery
Zhongang Cai, Mingyuan Zhang, Jiawei Ren, Chen Wei, Daxuan Ren, Zhengyu Lin, Haiyu Zhao, Lei Yang, Chen Change Loy, Ziwei Liu
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 |
|---|---|---|---|
| add GTA-Human by blended training | HMR · 3DPW test | PA-MPJPE (mm): 67.5 -> 60.5 (-7.0) | Improved |
| add GTA-Human by finetuning | HMR · 3DPW test | PA-MPJPE (mm): 67.5 -> 55.7 (-11.8) | Improved |
| add GTA-Human by finetuning | PARE · 3DPW test | PA-MPJPE (mm): 50.9 -> 46.8 (-4.1) | Improved |
| add GTA-Human to 3DPW training | VIBE · 3DPW test | PA-MPJPE (mm): 54.7 -> 52.4 | Improved |
| train on GTA-Human instead of AGORA | SPIN · 3DPW test | PA-MPJPE (mm): 55.3 -> 52.0 | Improved |
| add GTA-Human finetuning | SPIN · Human3.6M (Protocol 2) | PA-MPJPE (mm): 41.1 -> 40.8 | No clear effect |
| add Ganin et al. adversarial domain adaptation to blended training (ablation) | HMR · 3DPW test | PA-MPJPE (mm): 58.6 -> 55.5 | Improved |
| add CycleGAN sim2real augmentation to blended training (ablation) | HMR · 3DPW test | PA-MPJPE (mm): 58.6 -> 61.6 | Hurt |
| supervise with SMPL parameters instead of keypoints only (ablation) | HMR+ BT · 3DPW test | PA-MPJPE (mm): 60.9 -> 56.3 | Improved |
| increase GTA-Human data from 0 to 4x the real data | HMR+ BT · 3DPW test | PA-MPJPE (mm): 61.7 -> 56.0 | Improved |
| replace 25% of 300K real data with synthetic (75% real) | HMR+ BT · 3DPW test | PA-MPJPE (mm): 61.7 -> 56.8 | Improved |
| add GTA-Human to ResNet-152 backbone | HMR · 3DPW test | PA-MPJPE (mm): 58.4 -> 54.3 (-4.1) | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SPIN — SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition (AAAI 2021) · better on 2 of 2 shared results
- VIBE — VIBE: Video Inference for Human Body Pose and Shape Estimation (CVPR 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.86)
- PKU-DyMVHumans: A Multi-View Video Benchmark for High-Fidelity Dynamic Human Modeling (CVPR 2024 · similarity 0.85)
- Harmony4D: A Video Dataset for In-The-Wild Close Human Interactions (NeurIPS 2024 · similarity 0.85)
- BEDLAM2.0: Synthetic humans and cameras in motion (NeurIPS 2025 · similarity 0.85)
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.84)
- MVHumanNet: A Large-scale Dataset of Multi-view Daily Dressing Human Captures (CVPR 2024 · similarity 0.84)
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.
