FP: Emma Mohs, 2024

The Studio Ghibli anime style is known for its immersive realism and visual storytelling, which will be perfect for capturing my experience in Kentucky. I want to use this style to depict the gently rolling hills, horse pastures, and blossoming flowers that characterize the spring landscape in Kentucky. The scenery will be one of the most important parts of my video because I spent most of my time outside doing fieldwork in rural areas. The video will be a compilation of memories that take the viewer on a "day in the life" of doing field research - collecting pollen samples and counting flowers - while contemplating the relationship between humans and nature.

Using this model, I aim to explore the limitations and biases inherent in generative models, particularly their application for creating art and design. By varying the text prompts, I will test how the model responds to different inputs and observe how it interprets concepts and styles. The model's rapid feedback loop will help me identify the conditions under which the model performs best. It will also allow me to uncover biases related to representation, color, style, and subject matter, providing insights into the model's training data. By experimenting with various words and phrases, I will learn to write a prompt that gives my animation a uniform look and feel. Compared to other models we have used in class, the SDXL-Turbo will provide a hands-on approach to investigating generative models and their potential for creating artistic projects.