Institutional Repository of Vilnius University
Kelio atpažinimas dirbtiniais neuroniniais tinklais /
Abstract
dc:descriptionIn this thesis, the application of fully convolutional neural networks to road detection task of “KITTI” and “CityScape” dataset was investigated. The paper describes artificial neural networks theory, neural networks tendency to deepen and convolutional network basics. Next thesis describes chosen datasets and its preparation for training and verification of model and also describes chosen full convolutional networks “U-Net” and “OxfordNet” which are used for semantic segmentation. The paper describes conducted research of hyperparameters of implemented neural networks segmentizing “KITTI” dataset and later performance with all chosen datasets. Research results showed that implemented full convolutional networks can be used for some road detection tasks to solve.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Radzevičius, Laurynas,
- Contributors dc:contributor
-
- Masiulis, Rokas
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD107116896&prefLang=en_US