September 5, 2025 Dataset Open Access

Hussein Mohammed; Quang-Vinh Dang

Citation Style Language JSON Export

{"DOI":"10.25592/uhhfdm.18097","abstract":"<p>What&rsquo;s new in this version: This is the first complete version (v1.0). All images are in .png format, and the dataset annotations are divided into training (80 images)&nbsp;and test (20 images)&nbsp;sets.<br>\n<br>\nThis&nbsp;dataset is structured into four components, each serving a distinct role in the development of a&nbsp;document analysis system.&nbsp;</p>\n\n<ol>\n\t<li>\n\t<p><strong>Word-level annotations</strong> are provided in the file <code>word_annotations_for_cropped_images.json</code>. These annotations describe the images contained in the <code>cropped_images</code> folder. Each entry specifies the location of a word as a polygon, together with its orientation (horizontal, vertical, or tilted) and the type of writing implement used (ink or pencil). Additional metadata, such as bounding boxes and segmentation areas, is also included.</p>\n\t</li>\n\t<li>\n\t<p><strong>Cropped images</strong> are stored in the <code>cropped_images</code> folder. This set comprises 50 images, each containing only the primary page extracted from the corresponding full notebook scans.</p>\n\t</li>\n\t<li>\n\t<p><strong>Full images</strong> are located in the <code>full_images</code> folder. This collection also contains 50 items, representing the complete notebook scans in which the primary page appears alongside other material.</p>\n\t</li>\n\t<li>\n\t<p><strong>Page-level annotations</strong> are contained in the <code>page_annotations</code> folder. These are provided in YOLO format, with a single class (<code>page</code>) defined in <code>classes.txt</code>. Each annotation file specifies the bounding box of the primary page within the corresponding image in the <code>full_images</code> folder.</p>\n\t</li>\n</ol>\n\n<p>Examples illustrate the annotation structure. In the JSON file, a typical word annotation records polygon coordinates, the attribute <code>&quot;orientation&quot;: &quot;horizontal&quot;</code>, and <code>&quot;writing_tool&quot;: &quot;pencil&quot;</code>. In the YOLO annotations, a sample entry such as <code>0 0.499023 0.500776 0.777344 0.816912</code> denotes the normalised coordinates of the primary page bounding box.</p>\n\n<p><strong>Acknowledgement:</strong></p>\n\n<p>The research for this work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany&rsquo;s Excellence Strategy - EXC 2176 &lsquo;Understanding Written Artefacts: Material, Interaction and Transmission in Manuscript Cultures&rsquo;, project no. 390893796. The research was conducted within the scope of the Centre for the Study of Manuscript Cultures (CSMC) at Universit&auml;t Hamburg.<br>\n<br>\nThe images are taken from notebook pages of Rainer Maria Rilke, from the Deutsche Literaturarchiv Marbach (DLA), A:Rilke-Archiv Gernsbach.&nbsp;</p>\n\n<p>We thank Hui Xu for her support in annotating the images.</p>","author":[{"family":"Hussein Mohammed"},{"family":"Quang-Vinh Dang"}],"id":"18097","issued":{"date-parts":[[2025,9,5]]},"language":"eng","title":"The DLA-RMR dataset: Annotated subset of RMR notebooks for CVC development","type":"dataset","version":"1.0"}