Matching Items (4)
Filtering by
- All Subjects: Education
- All Subjects: semi-supervised learning
Description
Endowing machines with the ability to understand digital images is a critical task for a host of high-impact applications, including pathology detection in radiographic imaging, autonomous vehicles, and assistive technology for the visually impaired. Computer vision systems rely on large corpora of annotated data in order to train task-specific visual recognition models. Despite significant advances made over the past decade, the fact remains collecting and annotating the data needed to successfully train a model is a prohibitively expensive endeavor. Moreover, these models are prone to rapid performance degradation when applied to data sampled from a different domain. Recent works in the development of deep adaptation networks seek to overcome these challenges by facilitating transfer learning between source and target domains. In parallel, the unification of dominant semi-supervised learning techniques has illustrated unprecedented potential for utilizing unlabeled data to train classification models in defiance of discouragingly meager sets of annotated data.
In this thesis, a novel domain adaptation algorithm -- Domain Adaptive Fusion (DAF) -- is proposed, which encourages a domain-invariant linear relationship between the pixel-space of different domains and the prediction-space while being trained under a domain adversarial signal. The thoughtful combination of key components in unsupervised domain adaptation and semi-supervised learning enable DAF to effectively bridge the gap between source and target domains. Experiments performed on computer vision benchmark datasets for domain adaptation endorse the efficacy of this hybrid approach, outperforming all of the baseline architectures on most of the transfer tasks.
In this thesis, a novel domain adaptation algorithm -- Domain Adaptive Fusion (DAF) -- is proposed, which encourages a domain-invariant linear relationship between the pixel-space of different domains and the prediction-space while being trained under a domain adversarial signal. The thoughtful combination of key components in unsupervised domain adaptation and semi-supervised learning enable DAF to effectively bridge the gap between source and target domains. Experiments performed on computer vision benchmark datasets for domain adaptation endorse the efficacy of this hybrid approach, outperforming all of the baseline architectures on most of the transfer tasks.
ContributorsDudley, Andrew, M.S (Author) / Panchanathan, Sethuraman (Thesis advisor) / Venkateswara, Hemanth (Committee member) / McDaniel, Troy (Committee member) / Arizona State University (Publisher)
Created2019
Description
Humans have a great ability to recognize objects in different environments irrespective of their variations. However, the same does not apply to machine learning models which are unable to generalize to images of objects from different domains. The generalization of these models to new data is constrained by the domain gap. Many factors such as image background, image resolution, color, camera perspective and variations in the objects are responsible for the domain gap between the training data (source domain) and testing data (target domain). Domain adaptation algorithms aim to overcome the domain gap between the source and target domains and learn robust models that can perform well across both the domains.
This thesis provides solutions for the standard problem of unsupervised domain adaptation (UDA) and the more generic problem of generalized domain adaptation (GDA). The contributions of this thesis are as follows. (1) Certain and Consistent Domain Adaptation model for closed-set unsupervised domain adaptation by aligning the features of the source and target domain using deep neural networks. (2) A multi-adversarial deep learning model for generalized domain adaptation. (3) A gating model that detects out-of-distribution samples for generalized domain adaptation.
The models were tested across multiple computer vision datasets for domain adaptation.
The dissertation concludes with a discussion on the proposed approaches and future directions for research in closed set and generalized domain adaptation.
This thesis provides solutions for the standard problem of unsupervised domain adaptation (UDA) and the more generic problem of generalized domain adaptation (GDA). The contributions of this thesis are as follows. (1) Certain and Consistent Domain Adaptation model for closed-set unsupervised domain adaptation by aligning the features of the source and target domain using deep neural networks. (2) A multi-adversarial deep learning model for generalized domain adaptation. (3) A gating model that detects out-of-distribution samples for generalized domain adaptation.
The models were tested across multiple computer vision datasets for domain adaptation.
The dissertation concludes with a discussion on the proposed approaches and future directions for research in closed set and generalized domain adaptation.
ContributorsNagabandi, Bhadrinath (Author) / Panchanathan, Sethuraman (Thesis advisor) / Venkateswara, Hemanth (Thesis advisor) / McDaniel, Troy (Committee member) / Arizona State University (Publisher)
Created2020
Description
Although many data visualization diagrams can be made accessible for individuals who are blind or visually impaired, they often do not present the information in a way that intuitively allows readers to easily discern patterns in the data. In particular, accessible node graphs tend to use speech to describe the transitions between nodes. While the speech is easy to understand, readers can be overwhelmed by too much speech and may not be able to discern any structural patterns which occur in the graphs. Considering these limitations, this research seeks to find ways to better present transitions in node graphs.
This study aims to gain knowledge on how sequence patterns in node graphs can be perceived through speech and nonspeech audio. Users listened to short audio clips describing a sequence of transitions occurring in a node graph. User study results were evaluated based on accuracy and user feedback. Five common techniques were identified through the study, and the results will be used to help design a node graph tool to improve accessibility of node graph creation and exploration for individuals that are blind or visually impaired.
This study aims to gain knowledge on how sequence patterns in node graphs can be perceived through speech and nonspeech audio. Users listened to short audio clips describing a sequence of transitions occurring in a node graph. User study results were evaluated based on accuracy and user feedback. Five common techniques were identified through the study, and the results will be used to help design a node graph tool to improve accessibility of node graph creation and exploration for individuals that are blind or visually impaired.
ContributorsDarmawaskita, Nicole (Author) / McDaniel, Troy (Thesis director) / Duarte, Bryan (Committee member) / Computer Science and Engineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-12
Description
In this project, I investigated the impact of virtual reality on memory retention. The investigative approach to see the impact of virtual reality on memory retention, I utilized the memorization technique called the memory palace in a virtual reality environment. For the experiment, due to Covid-19, I was forced to be the only subject. To get effective data, I tested myself within randomly generated environments with a completely unique set of objects, both outside of a virtual reality environment and within one. First I conducted a set of 10 tests on myself by going through a virtual environment on my laptop and recalling as many objects I could within that environment. I recorded the accuracy of my own recollection as well as how long it took me to get through the data. Next I conducted a set of 10 tests on myself by going through the same virtual environment, but this time with an immersive virtual reality(VR) headset and a completely new set of objects. At the start of the project it was hypothesized that virtual reality would result in a higher memory retention rate versus simply going through the environment in a non-immersive environment. In the end, the results, albeit with a low test rate, leaned more toward showing the hypothesis to be true rather than not.
ContributorsDu, Michael Shan (Author) / Kobayashi, Yoshihiro (Thesis director) / McDaniel, Troy (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05