VML-MOC

VML-MOC数据集是由本古里安大学计算机科学系创建的，包含30页来自多个手稿的文档图像，主要用于处理多方向和弯曲的手写文本行。数据集中的文本行具有0°至180°的倾斜范围和各种弧形。创建过程中，使用了半自动化的标注系统Aletheia进行标注，提供了三种形式的标注结果。该数据集主要应用于历史手稿的阅读简化，旨在解决传统文本行分割方法在处理非水平或非直线文本行时的不足。

VML-MOC dataset was developed by the Department of Computer Science, Ben-Gurion University of the Negev. It contains 30 document images sourced from multiple manuscripts, and is primarily designed for processing multi-oriented and curved handwritten text lines. The text lines in this dataset have an inclination range from 0° to 180° and various curved forms. During the dataset construction, the semi-automated annotation system Aletheia was utilized to produce three types of annotation results. This dataset is mainly applied to facilitate the reading of historical manuscripts, aiming to address the limitations of traditional text line segmentation methods when handling non-horizontal or non-straight text lines.