Semantic Segmentation with Deep Convolutional Neural Networks

Αλεξανδρόπουλος, Σταμάτης; Alexandropoulos, Stamatis

dc.contributor.author	Αλεξανδρόπουλος, Σταμάτης	el
dc.contributor.author	Alexandropoulos, Stamatis	en
dc.date.accessioned	2023-01-11T07:14:43Z
dc.date.available	2023-01-11T07:14:43Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/56577
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.24275
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	Computer Vision	en
dc.subject	Semantic Segmentation	en
dc.subject	Deep Convolutional Neural Networks	en
dc.subject	Autonomous Driving	en
dc.subject	Offset Vectors	en
dc.subject	Όραση Υπολογιστών	el
dc.subject	Σημασιολογική Κατάτμηση	el
dc.subject	Βαθιά Συνελικτικά Νευρωνικά Δίκτυα	el
dc.subject	Αυτόνομη Οδήγηση	el
dc.subject	Διανύσματα Μετατόπισης	el
dc.title	Semantic Segmentation with Deep Convolutional Neural Networks	en
dc.contributor.department	ΕΡΓΑΣΤΗΡΙΟ ΟΡΑΣΗΣ ΥΠΟΛΟΓΙΣΤΩΝ, ΕΠΙΚΟΙΝΩΝΙΑΣ ΛΟΓΟΥ ΚΑΙ ΕΠΕΞΕΡΓΑΣΙΑΣ ΣΗΜΑΤΩΝ	el
heal.type	bachelorThesis
heal.classification	Computer Vision	en
heal.language	el
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2022-10-10
heal.abstract	Semantic segmentation is one of the fundamental topics of computer vision. Specifically, it is the process of assigning a category to each pixel in an image. There are a number of applications in a variety of fields, such as Autonomous Driving, Robotics, and Medical Image Processing, where pixel-level labeling is critical. Deep Convolutional Neural Networks (DCNNs) have lately demonstrated state-of-the-art performance in high-level recognition tasks. As a result, such models may now be used in the above-mentioned cutting-edge applications. Most of the related works concentrate on architectural changes to the used networks in order to better combine global context aggregation with local detail preservation, and utilize a simple loss computed on individual pixels. Designing more complex losses that account for the structure contained in semantic labelings has gotten substantially less attention. The goal of this thesis is to investigate such priors for semantic segmentation and to use them in the supervision of state-of-the-art networks to get results that better reflect the regularity of genuine segmentations. Based on knowledge about the high regularity of real scenes, we propose a method for improving class predictions by learning to selectively exploit information from coplanar pixels. In particular, we introduce a prior which claims that for each pixel, there is a seed pixel which shares the same prediction with the former. As a result of this, we design a network with two heads. The first head generates pixel-level classes, whereas the second generates a dense offset vector field that identifies seed pixel positions. Seed pixels’ class predictions are then utilized to predict classes at each point. To account for possible deviations from precise local planarity, the resultant prediction is adaptively fused with the initial prediction from the first head using a learnt confidence map. The entire architecture is implemented on HRNetV2, a state-of-the-art model on Cityscapes dataset. The offset vector-based HRNetV2 was trained on both Cityscapes and ACDC datasets. We assess our method through extensive qualitative and quantitative experiments and ablation studies and compare it with recent state-of-the-art methods demonstrating its superiority and advantages. To sum up, we achieve better results than the initial model.	en
heal.advisorName	Μαραγκός, Πέτρος	el
heal.advisorName	Σακαρίδης, Χρήστος	el
heal.committeeMemberName	Ποταμιάνος, Γεράσιμος	el
heal.committeeMemberName	Μαραγκός, Πέτρος	el
heal.committeeMemberName	Ροντογιάννης, Αθανάσιος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών	el
heal.academicPublisherID	ntua
heal.numberOfPages	121 σ.	el
heal.fullTextAvailability	false