VTechWorks staff will be away for the Thanksgiving holiday beginning at noon on Wednesday, November 27, through Friday, November 29. We will resume normal operations on Monday, December 2. Thank you for your patience.
 

ImageSI: Interactive Deep Learning for Image Semantic Interaction

Simple item page
dc.contributor.authorLin, Jiayueen
dc.contributor.committeechairNorth, Christopher L.en
dc.contributor.committeememberFaust, Rebecca Janeen
dc.contributor.committeememberHuang, Lifuen
dc.contributor.departmentComputer Science and#38; Applicationsen
dc.date.accessioned2024-06-05T08:02:26Zen
dc.date.available2024-06-05T08:02:26Zen
dc.date.issued2024-06-04en
dc.description.abstractInteractive deep learning frameworks are crucial for effectively exploring and analyzing complex image datasets in visual analytics. However, existing approaches often face challenges related to inference accuracy and adaptability. To address these issues, we propose ImageSI, a framework integrating deep learning models with semantic interaction techniques for interactive image data analysis. Unlike traditional methods, ImageSI directly incorporates user feedback into the image model, updating underlying embeddings through customized loss functions, thereby enhancing the performance of dimension reduction tasks. We introduce three variations of ImageSI, ImageSI$_{text{MDS}^{-1}}$, prioritizing explicit pairwise relationships from user interaction, and ImageSI$_{text{DRTriplet}}$ and ImageSI$_{text{PHTriplet}}$, emphasizing clustering by defining groups of images based on user input. Through usage scenarios and quantitative analyses centered on algorithms, we demonstrate the superior performance of ImageSI$_{text{DRTriplet}}$ and ImageSI$_{text{MDS}^{-1}}$ in terms of inference accuracy and interaction efficiency. Moreover, ImageSI$_{text{PHTriplet}}$ shows competitive results. The baseline model, WMDS$^{-1}$, generally exhibits lower performance metrics.en
dc.description.abstractgeneralInteractive deep learning frameworks are crucial for effectively exploring and analyzing complex image datasets in visual analytics. However, existing approaches often face challenges related to inference accuracy and adaptability. To address these issues, we propose ImageSI, a framework integrating deep learning models with semantic interaction techniques for interactive image data analysis. Unlike traditional methods, ImageSI directly incorporates user feedback into the image model, updating underlying embeddings through customized loss functions, thereby enhancing the performance of dimension reduction tasks. We introduce three variations of ImageSI, ImageSI$_{text{MDS}^{-1}}$, prioritizing explicit pairwise relationships from user interaction, and ImageSI$_{text{DRTriplet}}$ and ImageSI$_{text{PHTriplet}}$, emphasizing clustering by defining groups of images based on user input. Through usage scenarios and quantitative analyses centered on algorithms, we demonstrate the superior performance of ImageSI$_{text{DRTriplet}}$ and ImageSI$_{text{MDS}^{-1}}$ in terms of inference accuracy and interaction efficiency. Moreover, ImageSI$_{text{PHTriplet}}$ shows competitive results. The baseline model, WMDS$^{-1}$, generally exhibits lower performance metrics.en
dc.description.degreeMaster of Scienceen
dc.format.mediumETDen
dc.identifier.othervt_gsexam:40777en
dc.identifier.urihttps://hdl.handle.net/10919/119283en
dc.language.isoenen
dc.publisherVirginia Techen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectSemantic Interactionen
dc.subjectDeep Learningen
dc.subjectDimension Reductionen
dc.subjectImagesen
dc.titleImageSI: Interactive Deep Learning for Image Semantic Interactionen
dc.typeThesisen
thesis.degree.disciplineComputer Science & Applicationsen
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen
thesis.degree.levelmastersen
thesis.degree.nameMaster of Scienceen

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Lin_J_T_2024.pdf
Size:
3.33 MB
Format:
Adobe Portable Document Format
Download

Collections

Masters Theses