Unsupervised Learning of Text Line Segmentation by Differentiating Coarse Patterns

Berat Kurar Barakat*, Ahmad Droby, Raid Saabni, Jihad El-Sana

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


Despite recent advances in the field of supervised deep learning for text line segmentation, unsupervised deep learning solutions are beginning to gain popularity. In this paper, we present an unsupervised deep learning method that embeds document image patches to a compact Euclidean space where distances correspond to a coarse text line pattern similarity. Once this space has been produced, text line segmentation can be easily implemented using standard techniques with the embedded feature vectors. To train the model, we extract random pairs of document image patches with the assumption that neighbour patches contain a similar coarse trend of text lines, whereas if one of them is rotated, they contain different coarse trends of text lines. Doing well on this task requires the model to learn to recognize the text lines and their salient parts. The benefit of our approach is zero manual labelling effort. We evaluate the method qualitatively and quantitatively on several variants of text line segmentation datasets to demonstrate its effectivity.

Original languageEnglish
Title of host publicationDocument Analysis and Recognition – ICDAR 2021 - 16th International Conference, Proceedings
EditorsJosep Lladós, Daniel Lopresti, Seiichi Uchida
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages15
ISBN (Print)9783030863302
StatePublished - 2021
Externally publishedYes
Event16th International Conference on Document Analysis and Recognition, ICDAR 2021 - Lausanne, Switzerland
Duration: 5 Sep 202110 Sep 2021

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12822 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference16th International Conference on Document Analysis and Recognition, ICDAR 2021


  • Text line detection
  • Text line extraction
  • Text line segmentation
  • Unsupervised deep learning


Dive into the research topics of 'Unsupervised Learning of Text Line Segmentation by Differentiating Coarse Patterns'. Together they form a unique fingerprint.

Cite this