Research map / 3D vision / Face and shape
PaNDaS: Learnable Shape Interpolation Modeling with Localized Control
Thomas Besnier, Emery Pierson, Sylvain Arguillere, Maks Ovsjanikov, Mohamed Daoudi
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 PaNDaS instead of ARAPReg/SMS latent interpolation | PaNDaS · DFAUST interpolation | MSE: 6.2 -> 4.1 (ARAPReg), 7.2 -> 4.1 (SMS) | Improved |
| use PaNDaS instead of NJF/ARAPReg on COMA sequences | PaNDaS · COMA interpolation | MSE: 0.27 -> 0.06 (NJF), 0.08 -> 0.06 (ARAPReg) | Improved |
| use PaNDaS instead of localized ARAP for partial deformation | PaNDaS · DFAUST partial interpolation | MSE vs Cosine distance: MSE 4.4 -> 3.4; Cosd 0.08 -> 0.09 | Mixed |
| use PaNDaS instead of NJF/ARAP on scan partial interpolation | PaNDaS · DFAUST scans partial interpolation | Chamfer distance: 6.67 -> 4.80 (NJF), 6.03 -> 4.80 (ARAP) | Improved |
| add a cosine-distance normal loss on Jacobian-derived normals (ablation) | PaNDaS · mid-interpolation prediction | deformation smoothness (qualitative) | Improved |
| use more complex feature fusion than simple concatenation (ablation) | PaNDaS · generalization to new poses and partial deformations | generalization | Hurt |
Closest papers by meaning
- 4Deform: Neural Surface Deformation for Robust Shape Interpolation (CVPR 2025 · similarity 0.87)
- Neural Shape Deformation Priors (NeurIPS 2022 · similarity 0.87)
- 3D-CODED: 3D Correspondences by Deep Deformation (ECCV 2018 · similarity 0.87)
- Spectral Meets Spatial: Harmonising 3D Shape Matching and Interpolation (CVPR 2024 · similarity 0.85)
- Implicit-ARAP: Efficient Handle-Guided Neural Field Deformation via Local Patch Meshing (NeurIPS 2025 · similarity 0.85)
- Neural Pose Representation Learning for Generating and Transferring Non-Rigid Object Poses (NeurIPS 2024 · similarity 0.84)
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.
