Generative adversarial network as a basis for intelligent model of imaging architectural objects based on textual description (Kemerovo, Tomsk)
https://doi.org/10.31675/1607-1859-2023-25-5-84-94
Abstract
Вывод: рассматриваемая автоматизирующая система позволит существенно сократить временные, человеческие и денежные ресурсы, требуемые для разработки проекта будущего здания.
Перенести в английский вариант
Due to the high technology integrated into a person's daily life (smart house), this topic is relevant. One of elements of generative adversarial network is robot vacuum cleaners of various surface. Difficulties caused by this technique largely depend on the environment in which it locates.
Purpose: The development of the convolutional neural network concept allowing real-time distinguishing between the building interior and exterior.
Practical implication: The proposed intelligent system can distinguish between the building interior and exterior, that will considerably improve the firmware performance of modern technology in both the domestic and industrial segments.
About the Authors
P. A. Pylov
Gorbachev Kuzbass State Technical University
Russian Federation
Russian Federation
Petr A. Pylov, Research Assistant
Kemerovo
A. V. Dyagileva
Gorbachev Kuzbass State Technical University
Russian Federation
Russian Federation
Anna V. Dyagileva, PhD, A/Professor
Kemerovo
E. A. Nikolaeva
Gorbachev Kuzbass State Technical University
Russian Federation
Russian Federation
Evgenija A. Nikolaeva, PhD, A/Professor
Kemerovo
R. V. Maitak
Gorbachev Kuzbass State Technical University
Russian Federation
Russian Federation
Roman V. Maitak, Graduate Student
Kemerovo
T. A. Shalygina
Tomsk State University of Architecture and Building
Russian Federation
Russian Federation
Tat'jana A. Shalygina, PhD, A/Professor
Tomsk
References
1. Campanario G. The urban sketching handbook: Architecture and cityscapes: Tips and techniques for drawing on location. Quarry Books. 2014. 112 p. ISBN 9781592539611.
2. Radford А., Srivastava A., Morcoç S.B. The elements of modern architecture: Understanding contemporary buildings. Thames & Hudson. 2014. 344 p. ISBN 978-0500023624.
3. Bengio Y. Practical recommendations for gradient-based training of deep architectures. doi: 10.48550/arXiv.1206.5533 (accessed August 28, 2023).
4. Wolohan J.T. Mastering large datasets with Python. Manning Press. 2020. 312 p. ISBN 9781617296239.
5. Nikolenko S., Kadurin A., Arkhangelskaya E. Deep learning. Immersion in the world of neural networks. Saint-Petersburg: Piter, 2018. 482 p. ISBN 978-5-496-02536-2. (In Russian)
6. Kopec D. Classic computer science problems in Python. Manning Shelter Island. 2019. 201 p. ISBN 9781617295980.
7. Pylov P.A., Maitak R.V., Dyagileva A.V. Development of intelligent systems for processing signals from pressure sensors. Moscow, Vologda: Infra-Engineering, 2023. 172 p. ISBN 978-5-9729-1594-1. (In Russian)
8. Abrahams S., Hafner D. Tensor flow for machine learning intelligence. Bleeding Edge Press. 2019. 245 p.
9. Aggarwal Ch. Neural networks and deep learning. Saint-Petersburg: Dialektika, 2020. 752 p. ISBN 978-5-907203-01-3. (Russian translation)
10. Kogalov Yu. Forensic experts have named the cause of the Notre Dame fire. Available: https://rg.ru/2019/04/17/kriminalisty-nazvali-prichinu-pozhara-v-notr-dame.html (accessed October 8, 2023).
11. The Cathedral of Notre Dame de Paris will be restored thanks to the Assassin's Creed game. Available: www.rosbalt.ru/world/2019/04/16/1776251.html (accessed October 8, 2023).
Review
For citations:
Pylov P.A., Dyagileva A.V., Nikolaeva E.A., Maitak R.V., Shalygina T.A. Generative adversarial network as a basis for intelligent model of imaging architectural objects based on textual description (Kemerovo, Tomsk). Vestnik Tomskogo gosudarstvennogo arkhitekturno-stroitel'nogo universiteta. JOURNAL of Construction and Architecture. 2023;25(5):84-94. (In Russ.) https://doi.org/10.31675/1607-1859-2023-25-5-84-94
Views:
192
Email this article 