ferdous-alam/GenCAD

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

3669 stars · 452 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

GenCAD is a research codebase for image-conditioned CAD model generation using transformer-based contrastive representations (CCIP) and diffusion priors, published in TMLR 2025. It includes training scripts, pretrained checkpoints, and inference pipelines built on Python and pythonocc-core.

Use cases

• generate CAD models from images
• train a contrastive CAD representation model
• run diffusion prior training for CAD generation
• reconstruct CAD geometry from sketch images
• reproduce TMLR paper results on image-to-CAD generation

When to choose

• you need image-to-CAD generation with pretrained models
• you want to reproduce or extend the GenCAD paper
• you need a research baseline for generative CAD modeling

When to avoid

• you need a production CAD tool with a GUI
• you want a stable, licensed library for commercial use (no license specified)
• you lack GPU resources for training or inference

Facets

library · maturity active

machine-learning deep-learning computer-vision image-processing graphics simulation machine-learning computer-vision graphics artificial-intelligence developer-tools python cad-generation diffusion-models contrastive-learning transformer image-to-cad research-code pythonocc generative-models linux docker gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ferdous-alam/GenCAD")