dc.contributor.author | Plastiras, George | en |
dc.contributor.author | Kyrkou, Christos | en |
dc.contributor.author | Theocharides, Theocharis | en |
dc.coverage.spatial | Eindhoven, Netherlands | en |
dc.creator | Plastiras, George | en |
dc.creator | Kyrkou, Christos | en |
dc.creator | Theocharides, Theocharis | en |
dc.date.accessioned | 2021-01-26T09:45:46Z | |
dc.date.available | 2021-01-26T09:45:46Z | |
dc.date.issued | 2018 | |
dc.identifier.isbn | 978-1-4503-6511-6 | |
dc.identifier.uri | http://gnosis.library.ucy.ac.cy/handle/7/63393 | |
dc.description.abstract | Many applications utilizing Unmanned Aerial Vehicles (UAVs) require the use of computer vision algorithms to analyze the information captured from their on-board camera. Recent advances in deep learning have made it possible to use single-shot Convolutional Neural Network (CNN) detection algorithms that process the input image to detect various objects of interest. To keep the computational demands low these neural networks typically operate on small image sizes which, however, makes it difficult to detect small objects. This is further emphasized when considering UAVs equipped with cameras where due to the viewing range, objects tend to appear relatively small. This paper therefore, explores the trade-offs involved when maintaining the resolution of the objects of interest by extracting smaller patches (tiles) from the larger input image and processing them using a neural network. Specifically, we introduce an attention mechanism to focus on detecting objects only in some of the tiles and a memory mechanism to keep track of information for tiles that are not processed. Through the analysis of different methods and experiments we show that by carefully selecting which tiles to process we can considerably improve the detection accuracy while maintaining comparable performance to CNNs that resize and process a single image which makes the proposed approach suitable for UAV applications. | en |
dc.publisher | Association for Computing Machinery | en |
dc.source | Proceedings of the 12th International Conference on Distributed Smart Cameras | en |
dc.source.uri | https://doi.org/10.1145/3243394.3243692 | |
dc.title | Efficient ConvNet-based Object Detection for Unmanned Aerial Vehicles by Selective Tile Processing | en |
dc.type | info:eu-repo/semantics/conferenceObject | |
dc.identifier.doi | 10.1145/3243394.3243692 | |
dc.author.faculty | Πολυτεχνική Σχολή / Faculty of Engineering | |
dc.author.department | Τμήμα Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών / Department of Electrical and Computer Engineering | |
dc.type.uhtype | Conference Object | en |
dc.contributor.orcid | Kyrkou, Christos [0000-0002-7926-7642] | |
dc.contributor.orcid | Theocharides, Theocharis [0000-0001-7222-9152] | |
dc.gnosis.orcid | 0000-0002-7926-7642 | |
dc.gnosis.orcid | 0000-0001-7222-9152 | |