SMOL-MapSeg: Show Me One Label as prompt
This addresses the problem of flexible, concept-aware segmentation for historical map analysis, offering a scalable solution for researchers and archivists, though it is incremental as it builds on existing SAM-based fine-tuning methods.
The paper tackles the challenge of segmenting historical maps, which have inconsistent visual styles, by proposing SMOL-MapSeg, a method that uses explicit image-label prompts to guide models, resulting in accurate segmentation of user-defined classes and outperforming baselines with strong generalization on minimal data.
Historical maps offer valuable insights into changes on Earth's surface but pose challenges for modern segmentation models due to inconsistent visual styles and symbols. While deep learning models such as UNet and pre-trained foundation models perform well in domains like autonomous driving and medical imaging, they struggle with the variability of historical maps, where similar concepts appear in diverse forms. To address this issue, we propose On-Need Declarative (OND) knowledge-based prompting, a method that provides explicit image-label pair prompts to guide models in linking visual patterns with semantic concepts. This enables users to define and segment target concepts on demand, supporting flexible, concept-aware segmentation. Our approach replaces the prompt encoder of the Segment Anything Model (SAM) with the OND prompting mechanism and fine-tunes it on historical maps, creating SMOL-MapSeg (Show Me One Label). Unlike existing SAM-based fine-tuning methods that are class-agnostic or restricted to fixed classes, SMOL-MapSeg supports class-aware segmentation across arbitrary datasets. Experiments show that SMOL-MapSeg accurately segments user-defined classes and substantially outperforms baseline models. Furthermore, it demonstrates strong generalization even with minimal training data, highlighting its potential for scalable and adaptable historical map analysis.