- This event has passed.
Envisioning Stories with Generative AI
What if your e-reader had an “illustrate” option to quickly depict any passage in a book as a visual scene? Generative AI can now readily transform content from one modality to another. This enables a story written in text form to be instantly transferred to visual form, potentially enhancing readers’ engagement with the story. In this talk, Dr. Roemmele discusses the research endeavor of automatically generating scene illustrations for text-based stories, outlining current AI techniques for this task and the challenges that persist. She will also demonstrate an example implementation of this proposed e-reader “illustrate” feature using accessible tools and model APIs.
About the Speaker
Melissa Roemmele is a research scientist exploring the use of AI and NLP technologies to augment human creativity. She is currently part of the Midjourney Storytelling Lab, a team of researchers developing tools for AI-supported storytelling. Her work focuses on how to evaluate and optimize language-based interactions with these tools. She holds a PhD in Computer Science from the University of Southern California, where her 2018 dissertation explored emerging applications of language modeling to creative writing assistance.
About the Open Source Architect Community
The Open Source Architect (OSA) Community is an invitation-only group for seasoned software architects who are passionate about open source technology. Request to join the OSA Community: https://forms.gle/7efbynVzYhhH2LCQ7
We review each application carefully. If it’s a fit, you’ll get an invite to join the space where it all happens. For the latest updates on all things open source, follow our public feed on LinkedIn.