Improving Object Detection and Classification from 3D LiDAR Scans

Student: Borzunov Aleksandr

Educational Programme: Data Science (Master)

Year of Graduation: 2020

In this paper, we study the possibility of adapting a neural network architecture for anchor-free object detection FoveaBox for the case of processing 3D point clouds. We consider the problem of 3D object detection for predefined classes, as well as searching for arbitrary moving objects on the scene. We demonstrate that the resulting anchor-free architecture does not lose much quality and allows to create smaller, faster and simpler models that are more robust in predicting the size of arbitrary dynamic objects.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses