Research map / 3D vision / Face and shape
NeuForm: Adaptive Overfitting for Neural Shape Editing
Connor Zhizhen Lin, Niloy Mitra, Gordon Wetzstein, Leonidas Guibas, Paul Guerrero
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 |
|---|---|---|---|
| adaptive blending of generalizable and overfitted networks vs Spaghetti | NeuForm · PartNet chairs reconstruction, all regions | Chamfer Distance x100: 0.859 -> 0.293 | Improved |
| adaptive blending vs Coalesce | NeuForm · PartNet chairs reconstruction, all regions | Chamfer Distance x100: 0.446 -> 0.293 | Improved |
| adaptive blending vs generalizable only (ablation) | NeuForm · PartNet chairs reconstruction, all regions | Chamfer Distance x100: 0.457 -> 0.293 | Improved |
| adaptive blending vs overfitted only (ablation) | NeuForm · PartNet chairs reconstruction, all regions | Chamfer Distance x100: 0.238 -> 0.293 | Hurt |
| adaptive blending vs overfitted only (ablation) | NeuForm · PartNet chairs reconstruction, all regions | SDF error: 2.471 -> 2.176 | Improved |
| adaptive blending vs generalizable only (ablation) | NeuForm · PartNet chairs reconstruction, joint regions | Chamfer Distance x100: 0.390 -> 0.253 | Improved |
| adaptive blending vs overfitted only (ablation) | NeuForm · PartNet chairs reconstruction, non-joint regions | Chamfer Distance x100: 0.157 -> 0.334 | Hurt |
| adaptive blending vs generalizable only (ablation) | NeuForm · PartNet chairs reconstruction, non-joint regions | Chamfer Distance x100: 0.523 -> 0.334 | Improved |
| adaptive blending vs generalizable only on edits (ablation) | NeuForm · PartNet chair edits, all regions | Chamfer Distance: 0.047 -> 0.036 | Improved |
| adaptive blending vs overfitted only on edits (ablation) | NeuForm · PartNet chair edits, all regions | Chamfer Distance: 0.065 -> 0.036 | Improved |
| adaptive blending vs generalizable only on edited joint regions (ablation) | NeuForm · PartNet chair edits, edited joint regions | Chamfer Distance: 0.052 -> 0.052 | No clear effect |
| adaptive blending vs generalizable only on edited joint regions (ablation) | NeuForm · PartNet chair edits, edited joint regions | EMD: 48.07 -> 49.72 | Hurt |
Closest papers by meaning
- Neural Parts: Learning Expressive 3D Shape Abstractions With Invertible Neural Networks (CVPR 2021 · similarity 0.86)
- Interactive NeRF Geometry Editing With Shape Priors (TPAMI 2023 · similarity 0.86)
- ShapeTalk: A Language Dataset and Framework for 3D Shape Edits and Deformations (CVPR 2023 · similarity 0.86)
- NeuroMorph: Unsupervised Shape Interpolation and Correspondence in One Go (CVPR 2021 · similarity 0.86)
- $\texttt{NAISR}$: A 3D Neural Additive Model for Interpretable Shape Representation (ICLR 2024 · similarity 0.86)
- UniSDF: Unifying Neural Representations for High-Fidelity 3D Reconstruction of Complex Scenes with Reflections (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.
