Date of Award

2024

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

Committee Chair

Vineetha Menon

Committee Member

Jerome Baudry

Committee Member

Jacob Hauenstein

Research Advisor

Vineetha Menon

Subject(s)

American Sign Language--Translating, Context-aware computing, Closed captioning, Artificial intelligence

Abstract

Since the advent of Artificial Intelligence (AI), and the explosion of Genera- tive AI, much work has been done around image captioning, providing wildly accurate captions from still images, and more recently videos [3]. Alt-Text has improved dra- matically with efforts to enhance context and sentiment captured during automated captioning. This thesis attempts to expand on a section on Natural Language Pro- cessing (NLP) and its relevance to promoting accessibility and context-awareness in the premise of the American Sign Language (ASL) application. This thesis explores what it means to give full context about a caption or image to the end user, includ- ing sentiments, and sentient features existing within the string of words. This thesis goes to great lengths to expand on accessibility features currently present in image- captioning. Is a user-reliant (dependent on the users to provide a caption for the image) or a generic machine-generated alt-text enough for full accessibility? Could adding a sign language, for example, American Sign Language (ASL), be crucial to giving users more information than what is conveyed in the collection of words called a caption? Most importantly, how can we make this image-caption translation more context-aware to capture the nuances that might be important for user understand- ing? Ultimately, the goal is to make a world where every voice is heard and every individual feels welcome, and this thesis is a starting step towards an AI revolution that could be instrumental in achieving that dream.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.