Research map / 3D vision / Human pose estimation
Accurate 3D Body Shape Regression Using Metric and Semantic Attributes
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 |
|---|---|---|---|
| Train with crowd-sourced linguistic shape attributes rather than relying only on image-fitting cues | 3DShape · HBW test | vtovHD (mm): 32 -> 21 | Helped |
| Train with crowd-sourced linguistic shape attributes rather than relying only on image-fitting cues | 3DShape · MMTS | Chest MAE (mm): 107 -> 64 | Helped |
| Use linguistic attributes in addition to height supervision (ablation) | 3DShape · HBW test | vtovHD (mm): 22 -> 20 | Helped |
| Use linguistic attributes in addition to circumference supervision (ablation) | 3DShape · HBW test | vtovHD (mm): 26 -> 22 | Helped |
| Use linguistic attributes in addition to height and circumference supervision (ablation) | 3DShape · HBW test | vtovHD (mm): 23 -> 20 | Helped |
| Use silhouettes as the main shape cue instead of attribute-based weak supervision | 3DShape versus Sengupta et al. · SSP-3D test | PVE-T-SC: 13.6 -> 19.2 | Hurt |
| Use attributes, height, weight, and circumferences as inputs to shape reconstruction (ablation) | AtoS · CMTS male | vtovHD (mm): 7.3 ± 2.6 -> 5.8 ± 2.0 | Helped |
| Use quadratic polynomial regression instead of linear regression for attribute-to-shape mapping (ablation) | AtoS · CMTS females | V2V mean: 10.83 ± 4.77 -> 10.58 ± 4.67 | Helped |
Closest papers by meaning
- Probabilistic 3D Human Shape and Pose Estimation From Multiple Unconstrained Images in the Wild (CVPR 2021 · similarity 0.90)
- End-to-End Recovery of Human Shape and Pose (CVPR 2018 · similarity 0.90)
- Unite the People: Closing the Loop Between 3D and 2D Human Representations (CVPR 2017 · similarity 0.90)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · similarity 0.88)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 2024 · similarity 0.87)
- SCULPT: Shape-Conditioned Unpaired Learning of Pose-dependent Clothed and Textured Human Meshes (CVPR 2024 · similarity 0.87)
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.