Using dynamic allocation on mapped memory to process large point clouds with the PCL library
Автор: Beliaevskii Kirill
Журнал: Известия Самарского научного центра Российской академии наук @izvestiya-ssc
Рубрика: Информатика, вычислительная техника и управление
Статья в выпуске: 1 т.22, 2020 года.
Бесплатный доступ
The article proposes a solution to the problem of processing point clouds whose size exceeds available RAM by using Point Cloud Library (PCL). The problem of lack of RAM is solved using the dynamic allocation mechanism on the mapped memory. The methods of introducing such a mechanism into the PCL library are analyzed, and the performance of processing point clouds and the amount of memory consumed using RAM or mapped (external) memory is evaluated.
Point cloud library, point cloud, memory consumption, dynamic allocation, memory mapping
Короткий адрес: https://sciup.org/148314217
IDR: 148314217