ICDAR2015 competition on smartphone document capture and OCR (SmartDoc)

Burie, Jean-Christophe; Chazalon, Joseph; Coustaty, Mickaël; Eskenazi, Sébastien; Luqman, Muhammad; Mehri, Maroua; Nayef, Nibal; Ogier, Jm.; Prum, Sophea; Rusiñol, Marçal

doi:10.1109/icdar.2015.7333943

Cited by 73 publications

(67 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The method proposed in this paper has been applied to binarize documents in the challenge #2 (Smartphone OCR) of "ICDAR 2015 Competition on Smartphone Document Capture and OCR (SmartDoc)" [27]. We ranked 2nd in this competition among 8 contestants, with a character accuracy of 95.85% (note that the winner has taken advantage of the redundancy of documents in the test set to correct each individual document).…”

Section: Applying the Methods To Document Binarizationmentioning

confidence: 99%

Morphology-based hierarchical representation with application to text segmentation in natural images

Huỳnh

Géraud³

2016

2016 23rd International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Abstract-Many text segmentation methods are elaborate and thus are not suitable to real-time implementation on mobile devices. Having an efficient and effective method, robust to noise, blur, or uneven illumination, is interesting due to the increasing number of mobile applications needing text extraction. We propose a hierarchical image representation, based on the morphological Laplace operator, which is used to give a robust text segmentation. This representation relies on several very sound theoretical tools; its computation eventually translates to a simple labeling algorithm, and for text segmentation and grouping, to an easy tree-based processing. We also show that this method can also be applied to document binarization, with the interesting feature of getting also reverse-video text.

show abstract

Section: Applying the Methods To Document Binarizationmentioning

confidence: 99%

Morphology-based hierarchical representation with application to text segmentation in natural images

Huỳnh

Géraud³

2016

2016 23rd International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

show abstract

“…Target shape and coordinates of the region of interest are a bit more debatable in the sense that a real application would need either to ask the user for those elements, or discover it by itself. However, there already are some existing solutions for detecting the outline [1] of the document under capture and recover its shape [13], which produce good results. We therefore think that we should focus on the subsequent stages of the pipeline, and we believe this legitimates the fact that we do not simply provide the raw video sequence, but also some extra elements which facilitate the evaluation without simplifying the actual problem.…”

Section: Task Definitionmentioning

confidence: 99%

“…The SmartDoc series is an ongoing effort aiming at evaluating the performance of document imaging solutions using smartphones, and encouraging progress in this direction. The first release, organized as a competition for ICDAR 2015 [1], was composed of two challenges. SmartDoc 2015 -Challenge 1 was about locating a document page (its corners) within video frames to simulate how some applications could assist the user by detecting the document in real time during the preview phase preceding the capture.…”

Section: Introductionmentioning

confidence: 99%

SmartDoc 2017 Video Capture: Mobile Document Acquisition in Video Mode

Chazalon

Gomez‐Krämer

Burie

et al. 2017

2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our test dataset is the SmartDOC database for document capture (challenge 1) [14]. It consists of six different document types coming from public databases and five document images per class.…”

Section: A Datasetmentioning

confidence: 99%

Improving document matching performance by local descriptor filtering

Chazalon¹,

Rusiñol²,

Ogier³

2015

2015 13th International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

In this paper we propose an effective method aimed at reducing the amount of local descriptors to be indexed in a document matching framework. In an off-line training stage, the matching between the model document and incoming images is computed retaining the local descriptors from the model that steadily produce good matches. We have evaluated this approach by using the ICDAR2015 SmartDOC dataset containing near 25 000 images from documents to be captured by a mobile device. We have tested the performance of this filtering step by using ORB and SIFT local detectors and descriptors. The results show an important gain both in quality of the final matching as well as in time and space requirements.

show abstract

ICDAR2015 competition on smartphone document capture and OCR (SmartDoc)

Cited by 73 publications

References 7 publications

Morphology-based hierarchical representation with application to text segmentation in natural images

Morphology-based hierarchical representation with application to text segmentation in natural images

SmartDoc 2017 Video Capture: Mobile Document Acquisition in Video Mode

Improving document matching performance by local descriptor filtering

Contact Info

Product

Resources

About