Research map / 3D vision / Face and shape
Learning Shape Primitives via Implicit Convexity Regularization
Xiaoyang Huang, Yi Zhang, Kai Chen, Teng Li, Wenjun Zhang, Bingbing Ni
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 |
|---|---|---|---|
| add implicit convexity regularization (ICR) to primitive SDFs (ablation) | multi-view primitive learning · ShapeNet / CO3D | decomposition semantics | Improved |
| add ICR to primitive SDFs (ablation) | multi-view primitive learning · ShapeNet planes (wings and tails) | reconstruction fidelity | Hurt |
| add ICR to thin objects such as chair legs (ablation) | multi-view primitive learning · ShapeNet chair/table | thin-structure reconstruction | Improved |
| lower ICR weight lambda to 1 on selected primitives (others at 100) (ablation) | ICR · plane tail, pistol trigger | Chamfer distance | Improved |
| constrain second-order directional derivatives instead of SDF values directly (ablation) | ICR · training stability | training stability | Improved |
| use softmax-weighted-sum soft minimum instead of hard minimum over primitive SDFs (ablation) | ICR · primitive combination | convergence | Improved |
| ICR primitives instead of BAENet in 3D-supervised setting | ICR · ShapeNet single-view recon, Airplane | IoU: 52.4 -> 64.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- UNISURF — UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction (ICCV 2021) · better on 1 of 4 shared results
- CvxNet — CvxNet: Learnable Convex Decomposition (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
- Learning to Infer Implicit Surfaces without 3D Supervision (NeurIPS 2019 · similarity 0.89)
- Implicit Geometric Regularization for Learning Shapes (ICML 2020 · similarity 0.89)
- Few-Shot Unsupervised Implicit Neural Shape Representation Learning with Spatial Adversaries (ICML 2024 · similarity 0.89)
- Local Deep Implicit Functions for 3D Shape (CVPR 2020 · similarity 0.88)
- 3DIAS: 3D Shape Reconstruction With Implicit Algebraic Surfaces (ICCV 2021 · similarity 0.88)
- LP-DIF: Learning Local Pattern-Specific Deep Implicit Function for 3D Objects and Scenes (CVPR 2023 · similarity 0.88)
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.
