Research map / Point clouds / Shape and surface completion
DeepCAD: A Deep Generative Network for Computer-Aided Design Models
Rundi Wu, Chang Xiao, Changxi Zheng
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 random command-sequence augmentation (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | median chamfer distance: 0.787 -> 0.752 | Improved |
| add random command-sequence augmentation (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | invalid ratio: 3.30 -> 2.72 | Improved |
| quantize continuous parameters instead of regressing them (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | median chamfer distance: 2.142 -> 0.787 | Improved |
| quantize continuous parameters instead of regressing them (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | invalid ratio: 4.32 -> 3.30 | Improved |
| use absolute curve positions instead of relative positions (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | median chamfer distance: 0.863 -> 0.787 | Improved |
| use absolute curve positions instead of relative positions (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | ACC_param: 97.66 -> 97.47 | Hurt |
| translate sketch origin to loop start instead of adding loop start to extrusion command (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | median chamfer distance: 0.792 -> 0.787 | No clear effect |
| specify arcs by sweep angle and flag instead of middle point (ablation) | DeepCAD autoencoder · DeepCAD test set (autoencoding) | median chamfer distance: 0.790 -> 0.787 | No clear effect |
| DeepCAD generation compared with l-GAN | DeepCAD · DeepCAD shape generation | JSD: 5.02 -> 3.76 | Improved |
| DeepCAD generation compared with l-GAN | DeepCAD · DeepCAD shape generation | MMD: 1.27 -> 1.45 | Hurt |
| add uniform noise to input point clouds | DeepCAD point cloud to CAD · Point cloud to CAD reconstruction | invalid ratio: 12.08 -> 13.82 | Hurt |
Closest papers by meaning
- Revisiting CAD Model Generation by Learning Raster Sketch (AAAI 2025 · similarity 0.90)
- Generating CAD Code with Vision-Language Models for 3D Designs (ICLR 2025 · similarity 0.89)
- CADDreamer: CAD Object Generation from Single-view Images (CVPR 2025 · similarity 0.89)
- GeoCAD: Local Geometry-Controllable CAD Generation with Large Language Models (NeurIPS 2025 · similarity 0.89)
- CADCrafter: Generating Computer-Aided Design Models from Unconstrained Images (CVPR 2025 · similarity 0.88)
- CAD-Llama: Leveraging Large Language Models for Computer-Aided Design Parametric 3D Model Generation (CVPR 2025 · 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.
