# You Actually Look Twice At it (YALTAi): using an object detection approach instead of region segmentation within the Kraken engine

arXiv: https://arxiv.org/abs/2207.11230 · HTML: https://mlsanity.com/paper/2207.11230/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 29 references that are in the map.
- First posted on arXiv 19 Jul 2022.

## Builds on (most cited first, 4 of 4)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis](https://mlsanity.com/paper/2206.01062/) (2022)
- [A Two-Stage Method for Text Line Detection in Historical Documents](https://mlsanity.com/paper/1802.03345/) (2018)
- [BADAM: A Public Dataset for Baseline Detection in Arabic-script Manuscripts](https://mlsanity.com/paper/1907.04041/) (2019)

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

- [Multimodal LLMs for OCR, OCR Post-Correction, and Named Entity Recognition in Historical Documents](https://mlsanity.com/paper/2504.00414/) (2025)
- [Page Layout Analysis of Text-heavy Historical Documents: a Comparison of Textual and Visual Approaches](https://mlsanity.com/paper/2212.13924/) (2022)

## Abstract (arXiv)

Layout Analysis (the identification of zones and their classification) is the first step along line segmentation in Optical Character Recognition and similar tasks. The ability of identifying main body of text from marginal text or running titles makes the difference between extracting the work full text of a digitized book and noisy outputs. We show that most segmenters focus on pixel classification and that polygonization of this output has not been used as a target for the latest competition on historical document (ICDAR 2017 and onwards), despite being the focus in the early 2010s. We propose to shift, for efficiency, the task from a pixel classification-based polygonization to an object detection using isothetic rectangles. We compare the output of Kraken and YOLOv5 in terms of segmentation and show that the later severely outperforms the first on small datasets (1110 samples and below). We release two datasets for training and evaluation on historical documents as well as a new package, YALTAi, which injects YOLOv5 in the segmentation pipeline of Kraken 4.1.