Vision2CAD: A Visual Agent Harness for Explicit Geometry Referencing and Localization in Parametric CAD Modeling
TL;DR
Vision2CAD is presented, a visual agent harness that combines vision-language model (VLM) reasoning with deterministic CAD kernel operations and the Geometry Explicit Reference Dataset (GERD), which aligned commands, geometry states and IDs at every modeling step.
Abstract
Generating parametric CAD models requires accurate geometry and stable feature dependencies. Existing methods face challenges in selecting geometric references, interpreting sketch-plane local coordinates, and establishing sketch constraints to projected external geometry. We present Vision2CAD, a visual agent harness that combines vision-language model (VLM) reasoning with deterministic CAD kernel operations. An ID-based interface supports explicit geometry selection, a local-coordinate bridge converts view coordinates into sketch coordinates, and projected-edge localization supports external sketch constraints. These mechanisms establish feature dependencies within the supported modeling operations and constraint types. We also introduce the Geometry Explicit Reference Dataset (GERD), which aligned commands, geometry states and IDs at every modeling step. On GERD-EVL and a DeepCAD test subset, Vision2CAD improves mIoU by 11.1\% and 5.6\% and reduces Chamfer distance by 17.3\% and 41.8\%, respectively. Parameter-editing experiments and ablation studies further proved the preservation of parametric dependencies.
