# Importance of Textlines in Historical Document Classification

arXiv: https://arxiv.org/abs/2201.09575 · HTML: https://mlsanity.com/paper/2201.09575/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 9 references that are in the map.
- First posted on arXiv 24 Jan 2022.

## Builds on (most cited first, 5 of 5)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [Page Layout Analysis System for Unconstrained Historic Documents](https://mlsanity.com/paper/2102.11838/) (2021)
- [Convolutional Neural Networks for Font Classification](https://mlsanity.com/paper/1708.03669/) (2017)
- [Deep Generalized Max Pooling](https://mlsanity.com/paper/1908.05040/) (2019)

## Built on by (most cited first, 2 of 2)

- [A Survey of Historical Document Image Datasets](https://mlsanity.com/paper/2203.08504/) (2022)
- [Large Scale Genealogical Information Extraction From Handwritten Quebec Parish Records](https://mlsanity.com/paper/2304.14044/) (2023)

## Abstract (arXiv)

This paper describes a system prepared at Brno University of Technology for ICDAR 2021 Competition on Historical Document Classification, experiments leading to its design, and the main findings. The solved tasks include script and font classification, document origin localization, and dating. We combined patch-level and line-level approaches, where the line-level system utilizes an existing, publicly available page layout analysis engine. In both systems, neural networks provide local predictions which are combined into page-level decisions, and the results of both systems are fused using linear or log-linear interpolation. We propose loss functions suitable for weakly supervised classification problem where multiple possible labels are provided, and we propose loss functions suitable for interval regression in the dating task. The line-level system significantly improves results in script and font classification and in the dating task. The full system achieved 98.48 %, 88.84 %, and 79.69 % accuracy in the font, script, and location classification tasks respectively. In the dating task, our system achieved a mean absolute error of 21.91 years. Our system achieved the best results in all tasks and became the overall winner of the competition.