Research map / 3D vision / Human pose estimation
Learning to Fit Morphable Models
Vasileios Choutas, Federica Bogo, Jingjing Shen, Julien Valentin
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 |
|---|---|---|---|
| replace LGD update rule with the proposed weighted gradient/network update | LGD · 3DPW | PA-MPJPE (mm): 55.9 -> 53.9 | Improved |
| add GRU memory and architecture changes | Ours LGD + update rule · 3DPW | PA-MPJPE (mm): 53.9 -> 52.2 | Improved |
| add gradient descent term to network update (ablation) | Neural fitter · SMPL-H HMD ablation (half-space) | V2V head (mm): 14.7 -> 3.2 | Improved |
| add gradient descent term to network update (ablation) | Neural fitter · SMPL-H HMD ablation (half-space) | V2V full body (mm): 53.8 -> 49.9 | Improved |
| use per-step network weights instead of shared (ablation) | Neural fitter · SMPL-H HMD ablation (half-space) | V2V full body (mm): 52.3 -> 49.9 | Improved |
| use per-step network weights instead of shared (ablation) | Neural fitter · SMPL-H HMD ablation (half-space) | ground penetration (mm): 18.2 -> 15.5 | Improved |
| use GRU instead of ResNet feed-forward network (ablation) | Neural fitter · SMPL-H HMD ablation (half-space) | V2V full body (mm): 65.3 -> 53.6 | Improved |
| add second GRU layer (ablation) | Neural fitter · SMPL-H HMD ablation (half-space) | V2V full body (mm): 53.6 -> 49.9 | Improved |
| learn the step size gamma instead of constant 1e-4 (ablation) | Neural fitter · SMPL-H HMD ablation (half-space) | V2V full body (mm): 51.9 -> 49.9 | Improved |
| use vector instead of scalar lambda and gamma (ablation) | Neural fitter · 3DPW | PA-MPJPE (mm): 52.8 -> 52.3 | Improved |
| share network for lambda and gamma (ablation) | Neural fitter · 3DPW | PA-MPJPE (mm): 52.3 -> 52.2 | Improved |
| replace GRU with LSTM (ablation) | Neural fitter · not reported | not reported | No clear effect |
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 1 of 1 shared result
- 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
- Human Body Model Fitting by Learned Gradient Descent (ECCV 2020 · similarity 0.89)
- NPMs: Neural Parametric Models for 3D Deformable Shapes (ICCV 2021 · similarity 0.88)
- Learning to Reconstruct 3D Human Pose and Shape via Model-Fitting in the Loop (ICCV 2019 · similarity 0.88)
- Learned Vertex Descent: A New Direction for 3D Human Model Fitting (ECCV 2022 · similarity 0.87)
- Estimating Ego-Body Pose from Doubly Sparse Egocentric Video Data (NeurIPS 2024 · similarity 0.86)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 2024 · similarity 0.85)
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.
