Research map / 3D vision / Face and shape
Unsupervised Primitive Discovery for Improved 3D Generative Modeling
Salman H. Khan, Yulan Guo, Munawar Hayat, Nick Barnes
Check 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 |
|---|---|---|---|
| factorize shape generation into primitive generation followed by 3D shape refinement | Primitive GAN · ModelNet10 | accuracy | Improved |
| factorize shape generation using primitive representations | Primitive GAN · IKEA | AP: 55.0 -> 60.4 | Improved |
| add volumetric occupancy cost (ablation) | primitive discovery CRF · ModelNet10 | recall: 72.8 -> 83.0 | Improved |
| add shape uniformity cost (ablation) | primitive discovery CRF · ModelNet10 | recall: 81.5 -> 83.0 | Improved |
| add primitive compactness cost (ablation) | primitive discovery CRF · ModelNet10 | recall: 77.2 -> 83.0 | Improved |
| add support cost (ablation) | primitive discovery CRF · ModelNet10 | recall: 82.9 -> 83.0 | Improved |
| add shape symmetry cost (ablation) | primitive discovery CRF · ModelNet10 | recall: 80.0 -> 83.0 | Improved |
| use pairwise and higher-order CRF potentials in addition to unary costs (ablation) | primitive discovery CRF · ModelNet10 | recall: 51.6 -> 83.0 | Improved |
Closest papers by meaning
- "NeuSDFusion: A Spatial-Aware Generative Model for 3D Shape Completion, Reconstruction, and Generation" (ECCV 2024 · similarity 0.90)
- 3D-PRNN: Generating Shape Primitives With Recurrent Neural Networks (ICCV 2017 · similarity 0.90)
- High-Fidelity 3D Textured Shapes Generation by Sparse Encoding and Adversarial Decoding (ECCV 2024 · similarity 0.89)
- 3DILG: Irregular Latent Grids for 3D Generative Modeling (NeurIPS 2022 · similarity 0.88)
- Make-A-Shape: a Ten-Million-scale 3D Shape Model (ICML 2024 · similarity 0.87)
- Enhancing Implicit Shape Generators Using Topological Regularizations (ICML 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.
