Detail of the student project

Topic:Detekce 3D objektů v mračnech bodů
Department:Katedra kybernetiky
Supervisor:Ing. Tomáš Petříček, Ph.D.
Announce as:Diplomová práce, Bakalářská práce, Semestrální projekt
Description:Consider the task of 3D object detection in lidar data (point clouds), possibly using RGB image as additional cues, where each detected object is represented by its class (class confidences) and oriented 3D bounding box.
Bibliography:[1] Guo, Yulan, et al. "Deep learning for 3d point clouds: A survey." IEEE transactions on pattern analysis and machine intelligence (2020).
[2] Paszke, Adam, et al. "Pytorch: An imperative style, high-performance deep learning library." arXiv preprint arXiv:1912.01703 (2019).
[3] DARPA SubT Virtual Competition Software. URL:
Responsible person: Petr Pošík