Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX.
The paper presents the methods of development and the results of research on the effectiveness of the seq2seq neural network architecture using Visual Attention mechanism to solve the im2latex problem. The essence of the task is to create a neural network capable of converting an image with mathemat...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
Educational institution «Belarusian State University of Informatics and Radioelectronics»
2022-01-01
|
| Series: | Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki |
| Subjects: | |
| Online Access: | https://doklady.bsuir.by/jour/article/view/3243 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849336086692626432 |
|---|---|
| author | P. A. Vyaznikov I. D. Kotilevets |
| author_facet | P. A. Vyaznikov I. D. Kotilevets |
| author_sort | P. A. Vyaznikov |
| collection | DOAJ |
| description | The paper presents the methods of development and the results of research on the effectiveness of the seq2seq neural network architecture using Visual Attention mechanism to solve the im2latex problem. The essence of the task is to create a neural network capable of converting an image with mathematical expressions into a similar expression in the LaTeX markup language. This problem belongs to the Image Captioning type: the neural network scans the image and, based on the extracted features, generates a description in natural language. The proposed solution uses the seq2seq architecture, which contains the Encoder and Decoder mechanisms, as well as Bahdanau Attention. A series of experiments was conducted on training and measuring the effectiveness of several neural network models. |
| format | Article |
| id | doaj-art-eae422b593bf4dfe868c097e41fa910a |
| institution | Kabale University |
| issn | 1729-7648 |
| language | Russian |
| publishDate | 2022-01-01 |
| publisher | Educational institution «Belarusian State University of Informatics and Radioelectronics» |
| record_format | Article |
| series | Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki |
| spelling | doaj-art-eae422b593bf4dfe868c097e41fa910a2025-08-20T03:45:06ZrusEducational institution «Belarusian State University of Informatics and Radioelectronics»Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki1729-76482022-01-01198404410.35596/1729-7648-2021-19-8-40-441761Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX.P. A. Vyaznikov0I. D. Kotilevets1MIREA – Russian Technological UniversityMIREA – Russian Technological UniversityThe paper presents the methods of development and the results of research on the effectiveness of the seq2seq neural network architecture using Visual Attention mechanism to solve the im2latex problem. The essence of the task is to create a neural network capable of converting an image with mathematical expressions into a similar expression in the LaTeX markup language. This problem belongs to the Image Captioning type: the neural network scans the image and, based on the extracted features, generates a description in natural language. The proposed solution uses the seq2seq architecture, which contains the Encoder and Decoder mechanisms, as well as Bahdanau Attention. A series of experiments was conducted on training and measuring the effectiveness of several neural network models.https://doklady.bsuir.by/jour/article/view/3243im2latexseq2seqnlpneural network |
| spellingShingle | P. A. Vyaznikov I. D. Kotilevets Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX. Doklady Belorusskogo gosudarstvennogo universiteta informatiki i radioèlektroniki im2latex seq2seq nlp neural network |
| title | Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX. |
| title_full | Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX. |
| title_fullStr | Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX. |
| title_full_unstemmed | Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX. |
| title_short | Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX. |
| title_sort | developing a seq2seq neural network using visual attention to transform mathematical expressions from images to latex |
| topic | im2latex seq2seq nlp neural network |
| url | https://doklady.bsuir.by/jour/article/view/3243 |
| work_keys_str_mv | AT pavyaznikov developingaseq2seqneuralnetworkusingvisualattentiontotransformmathematicalexpressionsfromimagestolatex AT idkotilevets developingaseq2seqneuralnetworkusingvisualattentiontotransformmathematicalexpressionsfromimagestolatex |