Research map / Point clouds / Shape and surface completion
Generalized Deep 3D Shape Prior via Part-Discretized Diffusion Process
Read the paperCheck 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 |
|---|---|---|---|
| Use 3DQD instead of PVD | Shape generation · ShapeNet Airplane | CD: 73.82 -> 56.29 | Helped |
| Use 3DQD instead of PVD | Shape generation · ShapeNet Chair | CD: 56.26 -> 55.61 | Helped |
| Use 3DQD instead of PVD | Shape generation · ShapeNet Car | CD: 54.55 -> 55.75 | Hurt |
| Use 3DQD instead of AutoSDF | Shape completion · ShapeNet Bottom Half | MMD: 0.3510 -> 0.2933 | Helped |
| Use 3DQD instead of AutoSDF | Shape completion · ShapeNet Bottom Half | TMD: 0.0466 -> 0.0478 | Helped |
| Use 3DQD instead of AutoSDF | Shape completion · ShapeNet Octant | MMD: 0.5720 -> 0.4690 | Helped |
| Use 3DQD instead of AutoSDF | Shape completion · ShapeNet Octant | TMD: 0.0826 -> 0.0960 | Helped |
| Use 3DQD instead of AutoSDF | Text-guided shape generation · Text-guided shape generation | TMD: 0.1302 -> 0.2795 | Helped |
| Add Multi-frequency Fusion Module (MFM) (ablation) | 3DQD shape generation · Shape completion | high-frequency noise | Helped |
| Use classifier-free guidance | 3DQD multi-category generation · Shape completion | sample diversity | Helped |
| Use 100-step DDPM sampling instead of 1000 steps | 3DQD · Shape generation | inference steps: 100-step DDPM-based sampling; 1000 steps | Helped |
Closest papers by meaning
- GPLD3D: Latent Diffusion of 3D Shape Generative Models by Enforcing Geometric and Physical Priors (CVPR 2024 · similarity 0.88)
- ShapeShifter: 3D Variations Using Multiscale and Sparse Point-Voxel Diffusion (CVPR 2025 · similarity 0.87)
- Dora: Sampling and Benchmarking for 3D Shape Variational Auto-Encoders (CVPR 2025 · similarity 0.86)
- Sin3DM: Learning a Diffusion Model from a Single 3D Textured Shape (ICLR 2024 · similarity 0.86)
- LaGeM: A Large Geometry Model for 3D Representation Learning and Diffusion (ICLR 2025 · similarity 0.86)
- Flow-Corrected Shape Optimization: Taming Manifold Drift in High-Dimensional 3D Models (NeurIPS 2026 · 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.