Research map / 3D vision / Face and shape
Pop-Out Motion: 3D-Aware Image Deformation via Learning the Shape Laplacian
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 |
|---|---|---|---|
| learn the shape Laplacian rather than calculate it from mesh reconstruction or a point-cloud Laplacian | Pop-Out Motion · DFAUST, 16 handles | Weight L1 (×100): 2.66 -> 2.10 | Helped |
| learn the shape Laplacian rather than calculate it from mesh reconstruction or a point-cloud Laplacian | Pop-Out Motion · DFAUST, 32 handles | Shape CD (×100): 1.26 -> 1.45 | Hurt |
| learn the shape Laplacian rather than calculate it from mesh reconstruction or a point-cloud Laplacian | Pop-Out Motion · DFAUST, 16 handles | Shape HD (×0.1): 0.42 -> 0.42 | No clear effect |
| learn deformation weights from the shape Laplacian rather than using APSS + Marching Cubes | Pop-Out Motion · SHREC'16 partial point clouds, 16 handles | Weight L1 (×100): 5.15 -> 4.47 | Helped |
| learn deformation weights from the shape Laplacian rather than use Marching Cubes reconstruction | Pop-Out Motion · DeepHuman, 32 handles | Weight L1 (×100): 3.09 -> 2.14 | Helped |
| add KNN-Based Point Pair Sampling (ablation) | Laplacian Learning Network · DFAUST | Weight L1 (×100): 2.79 -> 2.10 | Helped |
| add sparsity weight function α (ablation) | Laplacian Learning Network · DFAUST | Weight L1 (×100): 4.95 -> 2.10 | Helped |
| use absolute difference and element-wise multiplication as the two pairwise feature functions rather than element-wise multiplication for both (ablation) | Laplacian Learning Network · DFAUST | Weight L1 (×100): 3.54 -> 2.10 | Helped |
| use absolute difference and element-wise multiplication as the two pairwise feature functions rather than absolute difference for both (ablation) | Laplacian Learning Network · DFAUST | Weight L1 (×100): 2.35 -> 2.10 | Helped |
| learn the shape Laplacian instead of directly learning handle-dependent deformation weights (ablation) | Pop-Out Motion · DFAUST, 16 handles | Weight L1 (×100): 9.08 -> 2.10 | Helped |
Closest papers by meaning
- Topologically-Aware Deformation Fields for Single-View 3D Reconstruction (CVPR 2022 · similarity 0.88)
- Unsupervised Learning of Dense Shape Correspondence (CVPR 2019 · similarity 0.87)
- Geometry-Aware Network for Non-Rigid Shape Prediction From a Single View (CVPR 2018 · similarity 0.87)
- Image Collection Pop-Up: 3D Reconstruction and Clustering of Rigid and Non-Rigid Categories (CVPR 2018 · similarity 0.86)
- 4Deform: Neural Surface Deformation for Robust Shape Interpolation (CVPR 2025 · similarity 0.86)
- Motion2VecSets: 4D Latent Vector Set Diffusion for Non-rigid Shape Reconstruction and Tracking (CVPR 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.