🤖 AI Summary
This work addresses the lack of publicly available datasets with executable ground-truth labels that has hindered reproducible evaluation in automatic conversion from hand-drawn sketches to UML diagrams. To bridge this gap, we present CAS2UML—the first open-source dataset comprising 557 manually annotated hand-drawn class and activity diagrams, each paired with verified, executable PlantUML code. Accompanying the dataset are validation tools and automated scripts that assess both syntactic correctness and renderability, enabling standardized and reproducible benchmarking. By providing high-quality annotations and essential evaluation infrastructure, this study fills a critical void in the field and establishes a foundational resource for future research on sketch-based UML diagram synthesis.
📝 Abstract
Automated UML generation from sketches and images is gaining renewed attention with the rise of large language models and multimodal AI. However, reproducible evaluation remains difficult due to the lack of public datasets with executable groundtruth models. We present CAS2UML, a public dataset of 557 handdrawn UML diagrams, including 271 class diagrams and 286 activity diagrams, each paired with manually validated PlantUML code. We also provide a PlantUML-based validation tool and reusable scripts for checking the syntactic correctness and renderability of generated UML artifacts, enabling reproducible benchmarking of sketch-to-UML approaches. The dataset, validation tool, processing scripts, documentation, and demonstration video are publicly available at: Dataset: https://huggingface.co/datasets/Seym0n /cas2uml_hand-drawn_to_plantuml_dataset; Tool and Scripts: https://github.com/Seym0n/handwritten-uml-dataset; Video: https://www.youtube.com/watch?v=KQrYeGgT3hs.