Detail of the student project

Topic:Hluboká neuronová síť pro detekci objektů za ztížených snímacích podmínek
Department:Katedra kybernetiky
Supervisor:Ing. Michal Reinštein, Ph.D.
Announce as:Diplomová práce, Bakalářská práce, Semestrální projekt
Description:The aim is to design and implement Deep Neural Network (DNN)[1, 2, 3] based solution for object detection in difficult environment of the DARPA SubT Challenge. Objects to be detected are specified in [4]; proprietary dataset is provided, including the labels. Understanding the dataset as well as the analysis of the state-of-the-art solutions in deep learning methods is integral part of the work.
Bibliography:[1] Goodfellow, Ian, et al. „Deep Learning“, MIT Press, 2016
[2] Tan, Mingxing, and Quoc V. Le. "Efficientnet: Rethinking model scaling for convolutional neural networks." arXiv preprint arXiv:1905.11946 (2019).
[3] Zhang, Aston, et al. "Dive into deep learning." Unpublished Draft. Retrieved 19 (2019): 2019.
[4] DARPA SubT Challenge
Responsible person: Petr Pošík