Research map / 3D vision / Face and shape
Sparse Representation for 3D Shape Estimation: A Convex Relaxation Approach
Xiaowei Zhou, Menglong Zhu, Spyridon Leonardos, Kostas Daniilidis
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 |
|---|---|---|---|
| use convex solution as initialization for alternating minimization instead of mean shape (ablation) | alternating minimization · CMU mocap human pose | 3D estimation error | Improved |
| add rotation synchronization and refinement after the convex solution (ablation) | Convex · FG3DCar with outliers | 2D shape error (pixels): 37.21 -> 34.76 | Improved |
| add rotation synchronization and refinement after the convex solution (ablation) | Convex · CMU mocap human pose | 3D estimation error | No clear effect |
| use convex relaxation instead of alternating minimization | shape and viewpoint estimation · FG3DCar with outliers | 2D shape error (pixels): 44.47 -> 37.21 | Improved |
| use convex relaxation instead of alternating minimization or PMP | shape and viewpoint estimation · CMU mocap human pose | 3D estimation error | Improved |
| add outlier term to the model (ablation) | convex and alternating models · CMU mocap with simulated outliers | 3D estimation error | Improved |
| use convex method instead of alternating under Gaussian noise | shape fitting · CMU mocap human pose with noise | 3D estimation error | Improved |
| use sparse coding instead of PCA for the shape dictionary (ablation) | pose dictionary · CMU mocap human pose | reconstruction error vs active bases | Improved |
| add nonnegativity constraint to sparse coding (ablation) | pose dictionary · CMU mocap human pose | representability | Hurt |
Closest papers by meaning
- In Perfect Shape: Certifiably Optimal 3D Shape Reconstruction From 2D Landmarks (CVPR 2020 · similarity 0.93)
- Revitalizing Optimization for 3D Human Pose and Shape Estimation: A Sparse Constrained Formulation (ICCV 2021 · similarity 0.89)
- Sparse-view Pose Estimation and Reconstruction via Analysis by Generative Synthesis (NeurIPS 2024 · similarity 0.89)
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (CVPR 2025 · similarity 0.87)
- TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation (CVPR 2024 · similarity 0.87)
- LEAP: Liberate Sparse-View 3D Modeling from Camera Poses (ICLR 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.
