Neural network technologies of recognition of three-dimensional objects

Бесплатный доступ

The article is devoted to the consideration of three-dimensional object recognition methods. Examples of the most famous convolutional neural network architectures used in the analysis of three-dimensional scenes are given. The main formats for representing three-dimensional data are listed. The results of testing the methods described in the article on a data set containing three-dimensional point clouds are presented.

Neural networks, three-dimensional point cloud, voxel network, object recognition, segmentation of objects

Короткий адрес: https://sciup.org/170185831

IDR: 170185831   |   DOI: 10.24411/2500-1000-2019-11889

Статья научная