Efficient and Lightweight Encoder-Decoder Architectures for Semantic Segmentation

dc.contributor.advisorFernandez, Amanda S.
dc.contributor.authorPerry, Jonathan
dc.contributor.committeeMemberGomez, Mauricio
dc.contributor.committeeMemberZhang, Weining
dc.creator.orcidhttps://orcid.org/0000−0002−1981−9740
dc.date.accessioned2024-02-12T19:30:26Z
dc.date.available2024-02-12T19:30:26Z
dc.date.issued2021
dc.descriptionThis item is available only to currently enrolled UTSA students, faculty or staff. To download, navigate to Log In in the top right-hand corner of this screen, then select Log in with my UTSA ID.
dc.description.abstractComputer Vision has benefited significantly from Deep Learning architectures, such as Convolutional Neural Networks (CNNs). Semantic Segmentation, which is pixel-wise image classification, is a key area of research within Computer Vision that has seen much growth from CNNs as well. With stand-alone embedded computer hardware becoming more capable and prominent, more applications of Semantic Segmentation have arisen. Thus, current technologies have been augmented, namely autonomous vehicles, and new technologies, including augmented reality (AR) and virtual reality (VR), have emerged in recent years. The effectiveness and performance of these technologies are dependent on object detection and, as a result, Semantic Segmentation has become a key aspect of research and development. CNNs designed for Semantic Segmentation tasks have shown significant performance in scene understanding, however most CNNs for these tasks require copious amounts of data as well as extensive computational resources. Consequently, current methods using CNNs for Semantic Segmentation tasks lack the real-time processing capabilities required for these technologies that are employed on embedded systems. In addition, the amount of annotated data readily available for AR/VR technologies is limited, rendering conventional training methods ineffective. In this work, a Minimized Efficient Network (MinENet) architecture is first shown to improve upon accuracy of Semantic Segmentation designed for an embedded AR/VR system. Second, this research presents EyeSeg, an encoder-decoder architecture, designed for accurate Semantic Segmentation with sparsely annotated data and applied to similar AR/VR problems. Lastly, CitySeg, an encoder-decoder architecture augmented with convolutional long-short term memory units (ConvLSTM), is presented as an extension of this work. CitySeg's preliminary results showcase the capabilities of efficient lightweight architectures while maintaining accuracy on data that has significantly higher dimensionality as well as much larger class feature size. This research reports results on OpenEDS2019 and OpenEDS2020 datasets compared against the related state-of-the-art approaches. Preliminary results on the CityScapes dataset are shown for the extended work of CitySeg using two different supervised training scenarios: utilizing CityScapes finely annotated training data. This work demonstrates real-time inference capabilities and accuracy performance in terms of mean Intersection over Union (mIoU), for embedded systems with limited memory and in scenarios that provide only sparsely annotated data.
dc.description.departmentComputer Science
dc.format.extent46 pages
dc.format.mimetypeapplication/pdf
dc.identifier.isbn9798516068140
dc.identifier.urihttps://hdl.handle.net/20.500.12588/4902
dc.languageen
dc.subjectAugmented reality
dc.subjectAutonomous vehicles
dc.subjectDeep learning
dc.subjectEye tracking
dc.subjectVirtual reality
dc.subjectComputer imaging
dc.subjectMinimized Efficient Network
dc.subjectReal-time
dc.subjectSemantic Segmentation
dc.subject.classificationComputer science
dc.subject.classificationArtificial intelligence
dc.subject.classificationComputer engineering
dc.titleEfficient and Lightweight Encoder-Decoder Architectures for Semantic Segmentation
dc.typeThesis
dc.type.dcmiText
dcterms.accessRightspq_closed
thesis.degree.departmentComputer Science
thesis.degree.grantorUniversity of Texas at San Antonio
thesis.degree.levelMasters
thesis.degree.nameMaster of Science

Files

Original bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Perry_utsa_1283M_13379.pdf
Size:
7.07 MB
Format:
Adobe Portable Document Format