IlluSonnet: Using Generative AI to Create Illustrations for Sonnets
Name
chen-tiffc-meng-eecs-2023-thesis.pdf
Description
Thesis PDF
Size
135.97 MB
Format
Adobe PDF
Checksum (MD5)
52c04b923888d7a51003e8f7d204d5b5
Author(s)
Chen, Tiffany
Advisor(s)
Mueller, Stefanie
Date Issued
September 2023
Publisher
Massachusetts Institute of Technology
Abstract
Poetry evokes imagery, and writers and readers alike desire to translate the artful wordplay to a beautiful image. To facilitate this process, we built IlluSonnet, a system that creates illustrations for poetry using text-to-image generative AI models. IlluSonnet works by labelling keywords, emotional qualities, and most related artistic style for the given sonnet before prompting DALL-E for an image. To evaluate IlluSonnet, we both ran a user study to assess the quality of the output images as well as the overall interface. Our study indicates that IlluSonnet helped users generate images that illustrated the sonnets well and that the process of creating and seeing imagery alongside the poem helped users understand the sonnets in a new light. We conclude by discussing how IlluSonnet can be used to further facilitate a deeper connection between both art and poetry.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
In Copyright - Educational Use Permitted
Copyright retained by author(s)
Persistent DSpace Link
