E.S. Budaev1, S.S. Mikhailova2, I.S. Evdokimova3, E.A. Khalmakshinov4

1,2 Financial University under the Government of the Russian Federation (Moscow, Russia)

3,4 East Siberian State University of Technology and Management (Ulan-Ude, Russia)

Problem setting. To date, image recognition on video continues to develop rapidly and provide new opportunities for creating intelligent systems in various fields. However, there are many problems associated with improving the adaptability, optimization, ethics and interpretability of models that require further research and development.

Target. Simplify and speed up the process of statistical analysis of social objects data received from the video stream.

Results. The study of existing algorithms and techniques, the selection of the most optimal of them, as well as the development and testing of a neural network model that can effectively and accurately recognize images in a video stream.

Practical significance. Analysis of various objects from video broadcasts, pre-processed video files for statistical accounting of these objects for various purposes. For example, maintaining statistics on the presence of people in a shopping center, the number of people passing on the street, etc.

Budaev E.S., Mikhailova S.S., Evdokimova I.S., Khalmakshinov E.A. Development of a neural network model for detecting objects in a video stream. Neurocomputers. 2023. V. 25. № 4. Р. 54-64. DOI: https://doi.org/10.18127/j19998554-202304-07 (In Russian)

1. Bishop K.M. Reinforcement learning. M.: Williams. 2016. 287 p. (in Russian)
2. Adamova A.A., Zaykin V.A., Gordeev D.V. Methods and technologies of machine learning in neural network for computer vision purposes. Neurocomputers. 2021. V. 23. № 4. Р. 25−39. DOI 10.18127/j19998554-202104-03. (in Russian)
3. Seliski R. Computer vision: algorithms and applications. M.: Springer. 2010. 812 p. ISBN 978-1848829343. (in Russian)
4. LeCun Y., Bengio Y., Hinton G. Deep Learning. Nature. 2015. V. 521. № 7553. P. 436–444. DOI 10.1038/nature14539.
5. Goodfellow I.Y., Bendjio I., Courville A. Deep learning. M.: MIPH. 2018. 736 p. ISBN 978-5-17-104617-4. (in Russian)
6. Guy V.E., Dmitriev A.V., Milov V.R., Vikulova E.N., Kuvshinov A.S., Nikiforov N.A. System of testing procedures for detecting and recognizing images in a video stream. Information-measuring and Control Systems. 2017. V. 15. № 8. P. 47–54. (in Russian)
7. Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., Gomez A.N., Kaiser L., Polosukhin I. Attention is All you Need. Neural Information Processing Systems. 2017. V. 30. P. 5998–6008. DOI 10.48550/arXiv.1706.03762.
8. He K., Zhang X., Ren S., Sun J. Deep Residual Learning for Image Recognition. IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA. 2016. P. 770–778. DOI 10.1109/CVPR.2016.90.
9. Redmon J., Divvala S., Girshick R., Farhadi A. You Only Look Once: Unified, Real-Time Object Detection. IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA. 2016. P. 779–788. DOI 10.1109/CVPR.2016.91.
10. Simonyan K., Zisserman A. Two-Stream Convolutional Networks for Action Recognition in Videos. Neural Information Processing Systems. 2014. V. 27. P. 568–576.
11. Sergienko A.B. Digital signal processing. St. Petersburg: Peter. 2017. 604 p. (in Russian).
12. Scholle F. Deep learning using Python. M.: DMK Press. 2019. 424 p. ISBN 978-5-97060-715-3. (in Russian)
13. Grineva N.V. Development of Strategy for Managing Efficiency of it Companies. Proceedings of 2018 11th International Conference "Management of Large-Scale System Development". Moscow: V.A. Trapeznikov Institute of Control SciencesMoscow. 2018.
    P. 8551864. DOI 10.1109/MLSD.2018.8551864.
14. Wang H., Kläser A., Schmid C., Liu C.L. Dense Trajectories and Motion Boundary Descriptors for Action Recognition. International Journal of Computer Vision. 2013. V. 103. № 1. P. 60–79. DOI 10.1007/s11263-012-0594-8.