Pu Li李朴

I work on 3D generative AI, with a focus on representations that connect learning-based methods with geometric design.

I received my Ph.D. from the Institute of Automation, Chinese Academy of Sciences (CASIA), advised by Prof. Dong-Ming Yan. I currently work at Huawei with Chen Yang and Jiemin Fang.

Google Scholar GitHub Email

Publications

2023–2025

Generated CAD shapes and their colorful Voronoi half-patch representations

SIGGRAPH ASIA 2025 · ACM TOG

BrepGPT: Autoregressive B-rep Generation with Voronoi Half-Patch

Pu Li^*, Wenhao Zhang^*, Weize Quan, Biao Zhang, Peter Wonka, Dong-Ming Yan

Paper · Project

A collection of gold-colored generated B-rep models

SIGGRAPH 2025

Stitch-A-Shape: Bottom-up Learning for B-Rep Generation

Pu Li^*, Wenhao Zhang^*, Jinglu Chen, Dong-Ming Yan

Paper

Raster sketches and generated CAD geometry from RECAD

AAAI 2025

Revisiting CAD Model Generation by Learning Raster Sketch

Pu Li^*, Wenhao Zhang^*, Jianwei Guo, Jinglu Chen, Dong-Ming Yan

Paper

Sketch-based CAD reconstruction and editable 3D shapes from SfmCAD

CVPR 2024

SfmCAD: Unsupervised CAD Reconstruction by Learning Sketch-based Feature Modeling Operations

Pu Li, Jianwei Guo, Huibin Li, Bedrich Benes, Dong-Ming Yan

Paper · Code

Deformed document images and their flattened results

ACM TOG 2023

Layout-aware Single-image Document Flattening

Pu Li, Weize Quan, Jianwei Guo, Dong-Ming Yan

Paper · Data

Self-supervised reconstruction of CAD models using sketch-extrude operations

CVPR 2023

SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operations

Pu Li, Jianwei Guo, Xiaopeng Zhang, Dong-Ming Yan

Paper · Code

∗ Equal contribution.

Resources

Awesome 3D Generation

A visual guide to 3D generative AI research, with curated papers across representations.

Awesome Neural CAD

A visual collection of neural CAD research, from generation and reconstruction to datasets.