细粒度历史书法风格理解数据集
收藏数据链接：
官方服务：
资源简介：
Addressing the "Knowledgeable but unperceptive" dilemma where existing Large Vision-Language Models (LVLMs) possess historical knowledge but lack fine-grained calligraphy style perception, we introduce HCSU—the first large-scale dataset and evaluation benchmark specifically tailored for fine-grained historical calligraphy style understanding. HCSU contains 39,307 meticulously annotated high-definition Chinese character images, accompanied by expert-level hierarchical aesthetic descriptions. Through a pioneering rigorous data processing pipeline, we successfully decouple authentic ink manuscripts (Tie) from stone rubbings (Bei), thoroughly resolving the "modality aliasing" problem that has long plagued the digital cultural heritage field.
提供机构：
maas创建时间：
2026-03-22
