In this research, we present a well-known problem in Computer vision (autonumous driving and driver assistant systems). From single monocular RGB image we designed a neural network to detect vehicles location, type and orientation. Our NN also provide an output mask for the bird view of a provided scene.

Али А. (науч. рук. Кашевник А.М.) Neural Network Model Architecture for 3D Vehicle Detection and Segmentation // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2022]. URL: https://kmu.itmo.ru/digests/article/8433