Description#
Storybuddy was developed as part of my master project during my studies at the Hochschule der Medien in Stuttgart. It was born out of the need to create storyboards quickly and efficiently, trying to harness the upcoming Text-to-Image AI technologies. The goal was to evaluate several Text-to-Image models and create a tool that would allow users to create storyboards with the help of AI. Just before the project was finished, DALL-E 2 was released, which was a game changer at the time, allowing for more complex and detailed images.
Technologies#
- PyTorch: Used in the backend to run the Text-to-Image models.
- Jina-Serve: The framework used to serve the models and handle requests.
- Next.js: The frontend framework used to build the web application.
- TailwindCSS: Used for styling the frontend.
Retrospection from Mid 2025#
As image generation has gotten a lot better since I worked on this project, especially when it comes to coherence between images, I would love to revisit this project. Maybe using ComfyUI workflows in the backend to allow for more complex image generation and editing.
Also, as I now have a better understanding of how to build web services, I would love to add all the features I wanted to add back then, like functioning user accounts, saving storyboards, and more.
