| DC Field | Value | Language |
|---|
| dc.contributor.author | Sholtanyuk, S. | - |
| dc.contributor.author | Leunikau, A. | - |
| dc.date.accessioned | 2021-11-08T12:03:11Z | - |
| dc.date.available | 2021-11-08T12:03:11Z | - |
| dc.date.issued | 2021 | - |
| dc.identifier.citation | Sholtanyuk, S. Lightweight Deep Neural Networks for Dense Crowd Counting Estimation / Sholtanyuk S., Leunikau A. // Pattern Recognition and Information Processing (PRIP'2021) = Распознавание образов и обработка информации (2021) : Proceedings of the 15th International Conference, 21–24 Sept. 2021, Minsk, Belarus / United Institute of Informatics Problems of the National Academy of Sciences of Belarus. – Minsk, 2021. – P. 61–64. | ru_RU |
| dc.identifier.uri | https://libeldoc.bsuir.by/handle/123456789/45869 | - |
| dc.description.abstract | In this paper, productiveness problems of deep neural networks for dense crowd counting prediction have been explored. Deep neural network CSRNet has been considered, and its shallow modifications (named CSRShNet-1 and CSRShNet-2) have been designed and researched. It has been shown that for relatively small crowds (up to 500 people) it is possible to reduce training time by using shallow networks with keeping an appropriate prediction accuracy. | ru_RU |
| dc.language.iso | en | ru_RU |
| dc.publisher | UIIP NASB | ru_RU |
| dc.subject | материалы конференций | ru_RU |
| dc.subject | conference proceedings | ru_RU |
| dc.subject | crowd counting | ru_RU |
| dc.subject | deep neural networks | ru_RU |
| dc.subject | convolutional neural networks | ru_RU |
| dc.subject | supervised learning | ru_RU |
| dc.subject | neural network performance | ru_RU |
| dc.subject | neural network accuracy | ru_RU |
| dc.title | Lightweight Deep Neural Networks for Dense Crowd Counting Estimation | ru_RU |
| dc.type | Статья | ru_RU |
| Appears in Collections: | Pattern Recognition and Information Processing (PRIP'2021) = Распознавание образов и обработка информации (2021)
|
