Research map / 3D vision / Face and shape
3D Human Mesh Regression With Dense Correspondence
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 DecoMR instead of SPIN | Mesh-based 3D human estimation · Human3.6M test | MPJPE-PA P2: 41.1 -> 39.3 | Helped |
| Use DecoMR instead of BodyNet | 3D human shape reconstruction · SURREAL test | Surface Error: 73.6 -> 56.5 | Helped |
| Use transferred local features instead of global features only (ablation) | DecoMR with SMPL UV map · Human3.6M test | MPJPE-PA P2: 49.1 -> 44.8 | Helped |
| Add transferred local features to global features (ablation) | DecoMR with SMPL UV map · Human3.6M test | MPJPE-PA P2: 49.1 -> 42.6 | Helped |
| Transfer raw image pixels instead of local image features (ablation) | DecoMR with SMPL UV map · Human3.6M test | MPJPE-PA P2: 42.6 -> 44.7 | Hurt |
| Use the continuous UV map instead of SMPL UV map (ablation) | DecoMR with global feature only · Human3.6M test | MPJPE-PA P2: 49.1 -> 47.1 | Helped |
| Use the continuous UV map instead of SMPL UV map (ablation) | DecoMR with local feature only · Human3.6M test | MPJPE-PA P2: 44.8 -> 42.3 | Helped |
| Use the continuous UV map instead of SMPL UV map (ablation) | DecoMR with global and local features · Human3.6M test | MPJPE-PA P2: 42.6 -> 39.3 | Helped |
| Use the continuous UV map instead of SMPL UV map (ablation) | DecoMR with global feature and transferred raw pixels · Human3.6M test | MPJPE-PA P2: 44.7 -> 42.6 | Helped |
| Use the continuous UV map instead of SMPL UV map (ablation) | UV-map representation · Original mesh and UV-map distance matrices | 2D correlation (S1): 0.2132 -> 0.7758 | Helped |
Closest papers by meaning
- MeshPose: Unifying DensePose and 3D Body Mesh Reconstruction (CVPR 2024 · similarity 0.89)
- GenHMR: Generative Human Mesh Recovery (AAAI 2025 · similarity 0.88)
- SAT-HMR: Real-Time Multi-Person 3D Mesh Estimation via Scale-Adaptive Tokens (CVPR 2025 · similarity 0.87)
- ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization (ICML 2025 · similarity 0.87)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · similarity 0.87)
- PostureHMR: Posture Transformation for 3D Human Mesh Recovery (CVPR 2024 · 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.