LLM-based CAD Information Extraction

Short description of the innovation study

The core idea of this proposal is to examine the impact of utilising generative AI, specifically multimodal Large Language Models (LLMs), to extract metadata from rendered 2D CAD (Computer-Aided Design) files within a data exchange solution for the automotive industry. CAD files are widely used in manufacturing sectors such as automotive, aerospace, and construction. Traditionally, these CAD files are archived as images, making it difficult to retrieve structured information. Current Optical Character Recognition (OCR) systems have limitations in accurately extracting tabular information and annotations from these images, particularly when dealing with complex layouts or dense information.

Organisations involved: 

End User: SSC-Services GmbH

ISV, Domain expert, HPC provider, Technology expert: The Karlsruhe Institute of Technology (KIT)