VML-MOC: Segmenting a Multiply Oriented and Curved Handwritten Text Line Dataset

Barakat, Berat Kurar; Cohen, Rafi; El-Sana, Jihad

doi:10.1109/icdarw.2019.50109

Cited by 7 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bounding polygon is used to enclose all the elements of a text line together with its neighbourhood background pixels [11,15]. Most of the extraction methods assume horizontally parallel text lines with constant heights, whereas some methods [5,2] are more generic.…”

Section: Related Workmentioning

confidence: 99%

“…However, this dataset is not enough representative of all the segmentation problems to evaluate a generic method. Hence, we also evaluated the proposed method on publicly available VML-MOC [5] dataset that contains multiply oriented and curved text lines with heterogeneous heights, and on VML-AHTE dataset that contains crowded diacritics.…”

Section: Related Workmentioning

confidence: 99%

“…1(a)). The proposed extraction method (FCN+EM) is evaluated on Visual Media Lab Arabic Handwritten Text line Extraction (VML-AHTE) dataset, Multiply Oriented and Curved (VML-MOC) dataset [5], and DIVA Historical Manuscript Database (DIVA-HisDB) [28]. VML-AHTE dataset is characterized by touching and overlapping text lines with close proximity and rich diacritics.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Text Line Extraction Using Fully Convolutional Network and Energy Minimization

Barakat

Droby

Alaasam

et al. 2021

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Text lines are important parts of handwritten document images and easier to analyze by further applications. Despite recent progress in text line detection, text line extraction from a handwritten document remains an unsolved task. This paper proposes to use a fully convolutional network for text line detection and energy minimization for text line extraction. Detected text lines are represented by blob lines that strike through the text lines. These blob lines assist an energy function for text line extraction. The detection stage can locate arbitrarily oriented text lines.Furthermore, the extraction stage is capable of finding out the pixels of text lines with various heights and interline proximity independent of their orientations. Besides, it can finely split the touching and overlapping text lines without an orientation assumption. We evaluate the proposed method on VML-AHTE, VML-MOC, and Diva-HisDB datasets. The VML-AHTE dataset contains overlapping, touching and close text lines with rich diacritics. The VML-MOC dataset is very challenging by its multiply oriented and skewed text lines. The Diva-HisDB dataset exhibits distinct text line heights and touching text lines. The results demonstrate the effectiveness of the method despite various types of challenges, yet using the same parameters in all the experiments.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Text Line Extraction Using Fully Convolutional Network and Energy Minimization

Barakat

Droby

Alaasam

et al. 2021

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, scarcity of labelled data causes rarely occurring curved text lines to be poorly detected. This problem has been handled via augmentation [9] or learning-free detection [13].…”

Section: Related Workmentioning

confidence: 99%

“…In this dataset we run our algorithm on presegmented main text regions by the given ground truth. The VML-MOC dataset [2] is characterized by multiply oriented and curved handwritten text lines.…”

Section: Datamentioning

confidence: 99%

Unsupervised Learning of Text Line Segmentation by Differentiating Coarse Patterns

Barakat

Droby

Saabni

et al. 2021

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

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.

show abstract