Automatic Detection of Pedestrians in Traffic Scene Images

Autor:	E. N. Yablokov, V. O. Yachnaya, V. R. Lutsiv, M. A. Mikhalkova
Rok vydání:	2020
Předmět:	Reduction (complexity) Artificial neural network Computer science Transfer of training business.industry Traffic scene Training (meteorology) Computer vision Artificial intelligence business Road traffic Task (project management)
Zdroj:	2020 Wave Electronics and its Application in Information and Telecommunication Systems (WECONF).
DOI:	10.1109/weconf48837.2020.9131458
Popis:	This article is devoted to automatic detection of pedestrians using a deep neural network. To achieve this goal, the deep neural network model DeepLabv3+ was adapted to the task of segmenting the road traffic scenes. The possibility of reduction of training sets by means of application of training transfer followed by fine tuning was of the primary interest.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::cb1e48a123b4d51d8b70da4f43d53528 https://doi.org/10.1109/weconf48837.2020.9131458 Zobrazit plný text záznamu