# BullingerDB: A Dataset for Handwritten Text Recognition and Writer Retrieval

arXiv: https://arxiv.org/abs/2605.30235 · HTML: https://mlsanity.com/paper/2605.30235/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 32 references that are in the map.
- First posted on arXiv 28 May 2026.

## Builds on (most cited first, 9 of 9)

- [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://mlsanity.com/paper/2109.10282/) (2021)
- [Unsupervised Feature Learning for Writer Identification and Writer Retrieval](https://mlsanity.com/paper/1705.09369/) (2017)
- [Best Practices for a Handwritten Text Recognition System](https://mlsanity.com/paper/2404.11339/) (2024)
- [ICDAR 2019 Competition on Image Retrieval for Historical Handwritten Documents](https://mlsanity.com/paper/1912.03713/) (2019)
- [Towards Writer Retrieval for Historical Datasets](https://mlsanity.com/paper/2305.05358/) (2023)
- [SAGHOG: Self-Supervised Autoencoder for Generating HOG Features for Writer Retrieval](https://mlsanity.com/paper/2404.17221/) (2024)
- [Self-Supervised Vision Transformers for Writer Retrieval](https://mlsanity.com/paper/2409.00751/) (2024)
- [Impact of Ground Truth Quality on Handwriting Recognition](https://mlsanity.com/paper/2312.09037/) (2023)
- [CTC Transcription Alignment of the Bullinger Letters: Automatic Improvement of Annotation Quality](https://mlsanity.com/paper/2508.07904/) (2025)

## Abstract (arXiv)

We present BullingerDB, a large-scale benchmark dataset for historical document analysis based on the correspondence of Heinrich Bullinger (1504-1575). The corpus comprises 20,898 pages and 499,222 text lines written by 796 writers over six decades, featuring stylistic variation, multilingual content (mostly Latin and Early New High German) as well as meta-information such as writer identity and time. We evaluate BullingerDB on text recognition and writer retrieval. TrOCR, the best performing model, achieves a CER of 9.1%. For writer retrieval, we introduce a temporal nDCG metric to assess time-aware retrieval. While temporally coherent retrieval is achievable, mAP (78.3%) scores indicate challenges due to long-term stylistic variation. With BullingerDB, we aim to establish a new benchmark for multilingual historical text recognition and temporally-aware writer analysis.