Title block detection and information extraction for enhanced building drawings search

2025-04-11

AI Summary

In the AEC industry, reliance on legacy architectural drawings is pervasive, yet non-standardized and noisy title blocks—especially in historical documents—hinder robust metadata extraction, impeding intelligent retrieval and regulatory compliance checking. To address this, we propose a lightweight CNN–GPT-4o collaborative reasoning framework that achieves high-robustness title block detection and structured metadata extraction. We introduce the first expert-annotated, scalable dataset supporting unified processing of both CAD vector drawings and hand-drawn sketches. Complementing this, we develop a domain-adaptive post-processing module and a semi-automatic annotation toolkit. Our method significantly outperforms state-of-the-art approaches on complex historical drawings, with marked improvements in detection and metadata extraction accuracy. Deployed in a production UI system, it boosts drawing retrieval efficiency by over 70%, enabling intelligent grouping, filtering, and automated compliance review.

Technology Category

• Natural Language Processing: Information Extraction
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Web Mining and Content Analysis: Bridging structured and unstructured data
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The architecture, engineering, and construction (AEC) industry still heavily relies on information stored in drawings for building construction, maintenance, compliance and error checks. However, information extraction (IE) from building drawings is often time-consuming and costly, especially when dealing with historical buildings. Drawing search can be simplified by leveraging the information stored in the title block portion of the drawing, which can be seen as drawing metadata. However, title block IE can be complex especially when dealing with historical drawings which do not follow existing standards for uniformity. This work performs a comparison of existing methods for this kind of IE task, and then proposes a novel title block detection and IE pipeline which outperforms existing methods, in particular when dealing with complex, noisy historical drawings. The pipeline is obtained by combining a lightweight Convolutional Neural Network and GPT-4o, the proposed inference pipeline detects building engineering title blocks with high accuracy, and then extract structured drawing metadata from the title blocks, which can be used for drawing search, filtering and grouping. The work demonstrates high accuracy and efficiency in IE for both vector (CAD) and hand-drawn (historical) drawings. A user interface (UI) that leverages the extracted metadata for drawing search is established and deployed on real projects, which demonstrates significant time savings. Additionally, an extensible domain-expert-annotated dataset for title block detection is developed, via an efficient AEC-friendly annotation workflow that lays the foundation for future work.

Problem

• Extracting metadata from building drawings for efficient search
• Improving title block detection in historical, non-standard drawings
• Combining CNN and GPT-4 for accurate information extraction

Innovation

• Combines CNN and GPT-4 for title block detection
• Extracts structured metadata from complex drawings
• Deploys efficient UI for drawing search