Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Filters
























31 Hits in 5.6 sec

HCR-Net: A deep learning based script independent handwritten character recognition network [article]

Vinod Kumar Chauhan, Sukhdeep Singh, Anuj Sharma
2024 arXiv   pre-print

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (794.7 kB)
https://web.archive.org/web/20230202054734/https://arxiv.org/pdf/2108.06663v3.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL. The file type is application/pdf.

Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:

2023 pre-print arXiv:2108.06663v3 fatcat:wfpxg2trkjem3ajz6zgutykkki

Other Versions

2021 pre-print
arXiv:2108.06663v1 fatcat:tkzocc6qi5e7rifpst7ckq2nxa
2022 pre-print
arXiv:2108.06663v2 fatcat:mxq6psh6gjf6xp74rr5ozegyki
Handwritten character recognition (HCR) remains a challenging pattern recognition problem despite decades of research, and lacks research on script independent recognition techniques.  ...  HCR-Net is extensively evaluated on 40 publicly available datasets of Bangla, Punjabi, Hindi, English, Swedish, Urdu, Farsi, Tibetan, Kannada, Malayalam, Telugu, Marathi, Nepali and Arabic languages, and  ...  Moreover, it is observed that a multi-column multi-scale CNN architecture proposed by [8] performs exceptionally well for Bangla script.  ... 
arXiv:2108.06663v4 fatcat:a5dof25aojhxpazmfjprtx5wca
Multiple Versions
Citation

Vinod Kumar Chauhan, Sukhdeep Singh, Anuj Sharma. "HCR-Net: A deep learning based script independent handwritten character recognition network." arXiv (2024)

Improving Scene Text Recognition for Indian Languages with Transfer Learning and Font Diversity

Sanjana Gunna, Rohit Saluja, Cheerakkuzhi Veluthemana Jawahar
2022 Journal of Imaging  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (2.7 MB)
https://web.archive.org/web/20220622004513/https://mdpi-res.com/d_attachment/jimaging/jimaging-08-00086/article_deploy/jimaging-08-00086-v2.pdf?version=1648114470

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL. The file type is application/pdf.

Further enriching the synthetic dataset with non-Unicode fonts and multiple augmentations helps us achieve a remarkable Word Recognition Rate gain of over 33% on the IIIT-ILST Hindi dataset.  ...  This work investigates the significant differences in Indian and Latin Scene Text Recognition (STR) systems.  ...  Data Availability Statement: Data available in a publicly accessible repository that does not issue DOIs.  ... 
doi:10.3390/jimaging8040086 pmid:35448213 pmcid:PMC9025185 fatcat:m7m6oppkevagblcx67yx3yku2u
DOAJ
Citation

Sanjana Gunna, Rohit Saluja, Cheerakkuzhi Veluthemana Jawahar. "Improving Scene Text Recognition for Indian Languages with Transfer Learning and Font Diversity." Journal of Imaging 8.4 (2022) 86

Bangla Sign Language Recognition and Sentence Building Using Deep Learning

Safayet Anowar Shurid, Khandaker Habibul Amin, Md. Shahnawaz Mirbahar, Dolan Karmaker, Mohammad Tanvir Mahtab, Farhan Tanvir Khan, Md. Golam Rabiul Alam, Md. Ashraful Alam
2020 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE)  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (2.8 MB)
https://web.archive.org/web/20210716154741/http://dspace.bracu.ac.bd/xmlui/bitstream/handle/10361/14455/16101030,%2016101093,%2016101025,%2016101058_CSE.pdf;jsessionid=CD9D8901669281A112132C71C8D18D82?sequence=1

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL. The file type is application/pdf.

This drove us towards finding a solution that would limit the gap instead of the communication and build a bridge instead of a boundary. ily communicate with them.  ...  Another paper that describes work on recognition and translation of sign languages, uses AI based solutions [6] .  ... 
doi:10.1109/csde50874.2020.9411523 fatcat:tfkrskin55hxpdzajiwtw2ooou
Citation

Safayet Anowar Shurid, Khandaker Habibul Amin, Md. Shahnawaz Mirbahar, Dolan Karmaker, Mohammad Tanvir Mahtab, Farhan Tanvir Khan, Md. Golam Rabiul Alam, Md. Ashraful Alam. "Bangla Sign Language Recognition and Sentence Building Using Deep Learning." 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE) (2020)

Derin Öğrenme Araştırma Alanlarının Literatür Taraması

M. Mutlu Yapıcı, Adem Tekerek, Nurettin Topaloğlu
2019 Gazi Mühendislik Bilimleri Dergisi  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (872.4 kB)
https://web.archive.org/web/20191231013541/https://dergipark.org.tr/tr/download/article-file/907912

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL. The file type is application/pdf.

Nowadays, many researchers try to find solutions to many problems in various fields under the light of DL methods.  ...  , voice and video recognition, medical image processing, and big data.  ...  [170] provided a head-to-head comparison between the latest technology in the mammography CAD system and the CNN to obtain a system that can read mammography independently.  ... 
doi:10.30855/gmbd.2019.03.01 fatcat:2sv7dg7elrfqppcjx5otzmb7pi
DOAJ
Citation

M. Mutlu Yapıcı, Adem Tekerek, Nurettin Topaloğlu. "Derin Öğrenme Araştırma Alanlarının Literatür Taraması." Gazi Mühendislik Bilimleri Dergisi 5.3 (2019) 188-215

Reading comprehension based question answering system in Bangla language with transformer-based learning

Tanjim Taharat Aurpa, Richita Khandakar Rifat, Md Shoaib Ahmed, Md. Musfique Anwar, A. B. M. Shawkat Ali
2022 Heliyon  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (888.1 kB)
https://web.archive.org/web/20221013030634/https://www.cell.com/heliyon/pdf/S2405-8440(22)02340-4.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL. The file type is application/pdf.

In the Bangla language, neither any dataset available for RC nor any work has been done in the past. In this research work, we develop a question-answering system from RC.  ...  For doing this, we construct a dataset containing 3636 reading comprehensions along with questions and answers.  ...  Multi-Head Attention: The multi-head self-attention mechanism in the transformer has three different uses.  ... 
doi:10.1016/j.heliyon.2022.e11052 pmid:36254291 pmcid:PMC9568857 fatcat:bgexsoawhrbb5isoeoo4ux3j7q
DOAJ Szczepanski
Citation

Tanjim Taharat Aurpa, Richita Khandakar Rifat, Md Shoaib Ahmed, Md. Musfique Anwar, A. B. M. Shawkat Ali. "Reading comprehension based question answering system in Bangla language with transformer-based learning." Heliyon 8.10 (2022) e11052

ICCIT 2020 Conference Proceedings [Front matter]

2020 2020 23rd International Conference on Computer and Information Technology (ICCIT)  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (3.1 MB)
https://web.archive.org/web/20210504034244/https://ieeexplore.ieee.org/ielx7/9392385/9392586/09392749.pdf?tp=&arnumber=9392749&isnumber=9392586&ref=

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL. The file type is application/pdf.

We further build a Bangla WordNet, BNNet, using these resources and map it with the Princeton English WordNet.  ...  With a variation in typical CNN model architecture, a satisfactory result is found to predict the genre of a particular artwork with an accuracy of 98.21%. 2 6 Omar Faruqe, Maliha Elma, Nahid Hossain  ...  Our Purpose is to build a platform with clustering algorithms which will jointly help to provide the quickest solution to find blood or plasma donor.  ... 
doi:10.1109/iccit51783.2020.9392749 fatcat:pz3hf7rsmzbjpe6hxjlu5tmrfq
Citation

"ICCIT 2020 Conference Proceedings [Front matter]." 2020 23rd International Conference on Computer and Information Technology (ICCIT) (2020)

Android Controlled Home Automation

2021 American Journal of Electronics & Communication  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (8.3 MB)
https://web.archive.org/web/20210617004626/https://ajec.smartsociety.org/wp-content/uploads/2021/06/AJEC_V1_IS3-1-1.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL. The file type is application/pdf.

The most used concept in planning is sampling-based planning, it provides a successful solution in wayfinding path planning, and because of this, it is performed in different robotics fields.  ...  The basic strategy of the overall behavior tasks in a behavior-based system are divided into smaller independent behaviors that focus on the performance of specific tasks such as the behavior of a robot  ...  From a statistical analysis we notice that the probability that a Bangla word will have at least one character with head-line is 0.994.  ... 
doi:10.15864/ajec.1303 fatcat:sitz5xykknhahawc54yptjptqa
Citation

"Android Controlled Home Automation." American Journal of Electronics & Communication 1.3 (2021)

Deep Learning for Sign Language Recognition: Current Techniques, Benchmarks, and Open Issues

Muhammad Al-Qurishi, Thariq Khalid, Riad Souissi
2021 IEEE Access  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (1.2 MB)
https://web.archive.org/web/20220522104844/https://ieeexplore.ieee.org/ielx7/6287639/6514899/09530569.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL. The file type is application/pdf.

In addition, recent advances have allowed researchers to move from simple recognition of sign language characters and words towards the capacity to translate continuous sign language communication with  ...  People with hearing impairments are found worldwide; therefore, the development of effective local level sign language recognition (SLR) tools is essential.  ...  [122] used the CNN approach to train a dataset obtained from the Bangla Sign Language.  ... 
doi:10.1109/access.2021.3110912 fatcat:mcjehb6znjcijhk2wzgdxbmzqq
DOAJ
Citation

Muhammad Al-Qurishi, Thariq Khalid, Riad Souissi. "Deep Learning for Sign Language Recognition: Current Techniques, Benchmarks, and Open Issues." IEEE Access (2021) 1-1

CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition [article]

Tianlun Zheng, Zhineng Chen, Shancheng Fang, Hongtao Xie, Yu-Gang Jiang
2023 arXiv   pre-print

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (3.3 MB)
https://web.archive.org/web/20231012192304/https://arxiv.org/pdf/2111.11011v4.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2023; you can also visit the original URL. The file type is application/pdf.

Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:

2023 pre-print arXiv:2111.11011v4 fatcat:uvnsrqy2affjtn7ih7y3ayxiti

Other Versions

2021 pre-print
arXiv:2111.11011v1 fatcat:lincjarqzna6jnvybxu6cq4kke
2021 pre-print
arXiv:2111.11011v2 fatcat:dsqxjfjh4fhr5m5jxq7mtkngza
2022 pre-print
arXiv:2111.11011v3 fatcat:55co5wh4zzdahg62ggcxiq4hiy
However, recent studies show that the two kinds of clues are not always well registered and therefore, feature and character might be misaligned in difficult text (e.g., with a rare shape).  ...  The Transformer-based encoder-decoder framework is becoming popular in scene text recognition, largely because it naturally integrates recognition clues from both visual and semantic domains.  ...  A recent study reported very fast yet efficient solutions by leveraging ViT-like architecture (Du et al., 2022) . We plan to incorporate it to speed up the recognition.  ... 
arXiv:2111.11011v5 fatcat:hdk2xpebr5fvzf7hwbov626ot4
Multiple Versions
Citation

Tianlun Zheng, Zhineng Chen, Shancheng Fang, Hongtao Xie, Yu-Gang Jiang. "CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition." arXiv (2023)

Intelligent Character Recognition System Using Convolutional Neural Network

S. Suriya, Dhivya S, Balaji M
2020 EAI Endorsed Transactions on Cloud Systems  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (4.6 MB)
https://web.archive.org/web/20210427175335/https://eudl.eu/pdf/10.4108/eai.16-10-2020.166659

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL. The file type is application/pdf.

Intellectual Character Recognition System is an application that uses Convolutional Neural Network (CNN) to recognize the Tamil character dataset accurately developed by HP Labs India.  ...  The proposed approach is capable of recognizing characters in a variety of challenging conditions using the Convolutional Neural Network, where traditional character recognition systems fail, notably in  ...  A Convolutional Neural Network (CNN, or ConvNet) is a unique sort of multi-layer neural systems, intended to perceive visual examples legitimately from pixel pictures with negligible preprocessing.  ... 
doi:10.4108/eai.16-10-2020.166659 fatcat:rrv3tyk2ezegdhcwsvuvvkgbrq
DOAJ
Citation

S. Suriya, Dhivya S, Balaji M. "Intelligent Character Recognition System Using Convolutional Neural Network." EAI Endorsed Transactions on Cloud Systems 6.19 (2020) 166659

International Conference on Emerging Technology in Computing, Communication and Electronics

2020 2020 Emerging Technology in Computing, Communication and Electronics (ETCCE)  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (3.1 MB)
https://web.archive.org/web/20210424044402/https://ieeexplore.ieee.org/ielx7/9350668/9350859/09350902.pdf?tp=&arnumber=9350902&isnumber=9350859&ref=

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL. The file type is application/pdf.

The author applies a most recent state-of-art scalingbased 3D-CNN different pretrained deep neural architectures, such as VGG19, and ResNet-50 to compare the performance with the proposed architecture.  ...  In this work, we have used multiple deep convolution neural networks (CNN) with the same architecture of InceptionV3.  ...  Paper ID: 121 Bangla Broadcast Speech Recognition Using Support Vector Machine Authors: Refat Noor Swarna Abstract: Over the past few decades, incredible growth has been revealed in the recognition  ... 
doi:10.1109/etcce51779.2020.9350902 fatcat:2drvbgkrmjabfeawuw33hui6yy
Citation

"International Conference on Emerging Technology in Computing, Communication and Electronics." 2020 Emerging Technology in Computing, Communication and Electronics (ETCCE) (2020)

Multi Task Deep Morphological Analyzer: Context Aware Joint Morphological Tagging and Lemma Prediction [article]

Saurav Jha, Akhilesh Sudhakar, Anil Kumar Singh
2019 arXiv   pre-print

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (2.4 MB)
https://web.archive.org/web/20200828141457/https://arxiv.org/ftp/arxiv/papers/1811/1811.08619.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL. The file type is application/pdf.

Other Versions

2018 pre-print
arXiv:1811.08619v1 fatcat:beudxf24nnhknb2vwcllvxzlom
We propose the Multi Task Deep Morphological analyzer (MT-DMA), a character-level neural morphological analyzer based on multitask learning of word-level tag markers for Hindi and Urdu.  ...  Exploiting character-level features in phonological space optimized for each tag using multi-objective genetic algorithm, our model establishes a new state-of-the-art accuracy score upon all seven of the  ...  This paper introduces a multi-task learning framework based upon two widely used architectures: (a) the aforementioned CNN-RNN model for predicting the POS, G, N, P, C, and TAM, and (b) an attention-based  ... 
arXiv:1811.08619v2 fatcat:qasqykysxjfm7oos4lkeg6mzze
Open Access Multiple Versions
Citation

Saurav Jha, Akhilesh Sudhakar, Anil Kumar Singh. "Multi Task Deep Morphological Analyzer: Context Aware Joint Morphological Tagging and Lemma Prediction." arXiv (2019)

uTHCD: A New Benchmarking for Tamil Handwritten OCR

Noushath Shaffi, Faizal Hajamohideen
2021 IEEE Access  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (6.4 MB)
https://web.archive.org/web/20210715104606/https://ieeexplore.ieee.org/ielx7/6287639/6514899/09481898.pdf?tp=&arnumber=9481898&isnumber=6514899&ref=

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL. The file type is application/pdf.

APPENDIX A HIERARCHICAL DATA FORMAT (HDF) FOR THE UTHDC DATABASE Listing 1. Python Code To Extract uTHCD Dataset  ...  BASIC CNN ARCHITECTURE We aim to design a minimalist CNN architecture to establish baseline accuracy for the proposed database.  ...  As this study predominantly uses the CNN architectures for reporting the use-case scenarios of the proposed dataset, the overall flow diagram of the model building using CNN is shown in Fig.12 .  ... 
doi:10.1109/access.2021.3096823 fatcat:rdrtnpzavzedtasnsc4ov4a23m
DOAJ
Citation

Noushath Shaffi, Faizal Hajamohideen. "uTHCD: A New Benchmarking for Tamil Handwritten OCR." IEEE Access (2021) 1-1

Survey on Aspect Category Detection

Siva Uday Sampreeth Chebolu, Paolo Rosso, Sudipta Kar, Thamar Solorio
2022 ACM Computing Surveys  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (2.0 MB)
https://web.archive.org/web/20220709065857/https://dl.acm.org/doi/pdf/10.1145/3544557

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL. The file type is application/pdf.

Various methods have been proposed to deal with such a problem.  ...  In this article, we first introduce several datasets in the community that deal with this task and take a closer look at them by providing some exploratory analysis.  ...  entity recognition, text summarization, and sentiment analysis.  ... 
doi:10.1145/3544557 fatcat:zrhucn4xbvguxgtfri7hd2ssdm
Citation

Siva Uday Sampreeth Chebolu, Paolo Rosso, Sudipta Kar, Thamar Solorio. "Survey on Aspect Category Detection." ACM Computing Surveys 55.7 (2022) 1-37

Transcription Alignment of Historical Vietnamese Manuscripts without Human-Annotated Learning Samples

Anna Scius-Bertrand, Michael Jungo, Beat Wolf, Andreas Fischer, Marc Bui
2021 Applied Sciences  

Preserved Fulltext

fulltext thumbnail
Web Archive Capture PDF (27.2 MB)
https://web.archive.org/web/20210604221238/https://res.mdpi.com/d_attachment/applsci/applsci-11-04894/article_deploy/applsci-11-04894-v2.pdf

A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL. The file type is application/pdf.

a detection accuracy of 96.4% with a YOLOv5m model without using any human annotation.  ...  Transcription alignment is a simpler task that aims to find a correspondence between text in the scanned image and its existing Unicode counterpart, a correspondence which can then be used as training  ...  Furthermore, we thank Kha Cong Nguyen, Cuong Tuan Nguyen, and Masaki Nakagawa, the authors of [3] , for providing us with their ground truth for the test images.  ... 
doi:10.3390/app11114894 fatcat:sfh27dqhgzawzbu7nndzgokfty
DOAJ
Citation

Anna Scius-Bertrand, Michael Jungo, Beat Wolf, Andreas Fischer, Marc Bui. "Transcription Alignment of Historical Vietnamese Manuscripts without Human-Annotated Learning Samples." Applied Sciences 11.11 (2021) 4894

« Previous Showing results 1 — 15 out of 31 results