DrawingsDreamer generates precise multi-view CAD drawings with AI
DrawingsDreamer: A Unified Multi-View Engineering Drawings Generation Model
Computer Vision and Pattern Recognition
Summary
Engineering drawings need to be very exact and show different views of the same object, which is hard for many AI tools designed for artistic images. The authors created DrawingsDreamer, a system that uses a special way of turning drawings into sequences so a language model can produce detailed and aligned vector drawings from multiple views. This approach avoids working with pixel images and focuses on the drawing instructions themselves. The model was trained step-by-step, starting from small fixes to generating entire drawings, and it performs well in both accuracy and correctness.
What this means in practice
- •For cad software developers: Integrate vector-based multi-view drawing generation into CAD tools to streamline creating aligned engineering drawings without relying on raster images.
- •For industrial design teams: Automatically generate precise multi-view engineering blueprints that adhere to geometric constraints and spatial alignment for complex parts.
Authors
Shurui Liu, Weide Chen, Changwang Yi, Ancong Wu
Abstract
Scalable Vector Graphics (SVG) are essential for modern industrial Computer-Aided Design (CAD). However, existing autoregressive SVG generation models are predominantly tailored for artistic creation and struggle to maintain the rigorous geometric fidelity and cross-view spatial alignment required for engineering drawings. To bridge this gap, we introduce \textbf{DrawingsDreamer}, a unified Large Language Model (LLM)-driven framework for multi-view vector-based engineering drawings generation. By formulating the generation of multi-view engineering drawings purely as a sequence modeling task, we eliminate the need of raster image encoders. We propose a Streamlined Representation utilizing hierarchical postfix tokenization, which guides the model to establish local geometric coordinates before assigning semantic boundaries. Optimized via a progressive task-aware curriculum schedule, \textbf{DrawingsDreamer} effectively transitions from localized structural repair to macroscopic generation in a unified model. Extensive experiments demonstrate that our unified model achieves strong performance in both geometric fidelity and syntactic accuracy across diverse conditional and unconditional generation tasks.