JoinABLe: Learning Bottom-up Assembly of Parametric CAD Joints

TL;DR

JoinABLe is a learning-based approach that assembles CAD parts into joints using weak supervision, achieving near-human accuracy and facilitating future research with a new dataset.

Contribution

The paper introduces JoinABLe, a novel method for assembling CAD parts into joints using graph-based learning without object labels or human guidance.

Findings

Achieves 79.53% accuracy in joint assembly

Outperforms multiple baseline methods

Releases a comprehensive CAD assembly dataset

Abstract

Physical products are often complex assemblies combining a multitude of 3D parts modeled in computer-aided design (CAD) software. CAD designers build up these assemblies by aligning individual parts to one another using constraints called joints. In this paper we introduce JoinABLe, a learning-based method that assembles parts together to form joints. JoinABLe uses the weak supervision available in standard parametric CAD files without the help of object class labels or human guidance. Our results show that by making network predictions over a graph representation of solid models we can outperform multiple baseline methods with an accuracy (79.53%) that approaches human performance (80%). Finally, to support future research we release the Fusion 360 Gallery assembly dataset, containing assemblies with rich information on joints, contact surfaces, holes, and the underlying assembly graph structure.