Parametric Primitive Analysis of CAD Sketches with Vision Transformer

TL;DR

This paper introduces a two-stage neural network framework using Vision Transformers for analyzing CAD sketches, effectively predicting primitives and constraints with improved accuracy and interpretability.

Contribution

It presents a novel two-stage network that decouples primitive types from parameters and explicitly models constraints, addressing error accumulation and complexity issues in CAD sketch analysis.

Findings

Outperforms existing methods on public datasets

Enhances interpretability with pointer modules

Reduces error propagation in sketch analysis

Abstract

The design and analysis of Computer-Aided Design (CAD) sketches play a crucial role in industrial product design, primarily involving CAD primitives and their inter-primitive constraints. To address challenges related to error accumulation in autoregressive models and the complexities associated with self-supervised model design for this task, we propose a two-stage network framework. This framework consists of a primitive network and a constraint network, transforming the sketch analysis task into a set prediction problem to enhance the effective handling of primitives and constraints. By decoupling target types from parameters, the model gains increased flexibility and optimization while reducing complexity. Additionally, the constraint network incorporates a pointer module to explicitly indicate the relationship between constraint parameters and primitive indices, enhancing interpretability and performance. Qualitative and quantitative analyses on two publicly available datasets demonstrate the superiority of this method.