Research map / 3D vision / Human pose estimation
Capturing Humans in Motion: Temporal-Attentive 3D Human Pose and Shape Estimation From Monocular Video
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 |
|---|---|---|---|
| Use MPS-Net instead of TCMR | 3DPW · 3DPW | MPJPE: 86.5 -> 84.3 | Helped |
| Use MPS-Net instead of TCMR | MPI-INF-3DHP · MPI-INF-3DHP | PA-MPJPE: 63.5 -> 62.8 | Helped |
| Use MPS-Net instead of TCMR | Human3.6M · Human3.6M | MPJPE: 73.6 -> 69.4 | Helped |
| Use MPS-Net instead of TCMR | 3DPW · 3DPW | ACC-ERR: 6.8 -> 7.4 | Hurt |
| Use MPS-Net instead of TCMR | MPI-INF-3DHP · MPI-INF-3DHP | ACC-ERR: 8.5 -> 9.6 | Hurt |
| Add NSSM-guided MoCA instead of using only a non-local block (ablation) | MPS-Net, without HAFI · 3DPW | MPJPE: 87.6 -> 86.7 | Helped |
| Add HAFI instead of leaving the model with MoCA only (ablation) | MPS-Net · 3DPW | MPJPE: 86.7 -> 84.3 | Helped |
| Use HAFI instead of temporal feature integration (ablation) | MPS-Net with MoCA · 3DPW | ACC-ERR: 10.5 -> 7.4 | Helped |
| Use 3 frames per HAFI group rather than 2 (ablation) | MPS-Net · 3DPW | MPJPE: 85.4 -> 84.3 | Helped |
| Use 3 frames per HAFI group rather than 4 (ablation) | MPS-Net · 3DPW | MPJPE: 85.9 -> 84.3 | Helped |
| Add HAFI to a MoCA-free model (ablation) | MPS-Net, HAFI only · 3DPW | MPJPE: 87.6; 86.7 | Mixed |
| Use MoCA and HAFI together rather than only HAFI (ablation) | MPS-Net · 3DPW | MPJPE: 87.6 -> 84.3 | Helped |
Closest papers by meaning
- Capturing the Motion of Every Joint: 3D Human Pose and Shape Estimation with Independent Tokens (ICLR 2023 · similarity 0.92)
- 3D Human Pose Estimation Using Spatio-Temporal Networks with Explicit Occlusion Training (AAAI 2020 · similarity 0.92)
- Motions as Queries: One-Stage Multi-Person Holistic Human Motion Capture (CVPR 2025 · similarity 0.90)
- Optimizing Human Pose Estimation Through Focused Human and Joint Regions (AAAI 2025 · similarity 0.89)
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.89)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · 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.