Role
AI has benefitted our project in multiple ways. Thanks to image generation AI, a lot of work was done in the blink of an eye. Specifically image generation and modification has been made much easier allowing our desired images to be created given our skillsets. We developed our understanding of making strong prompts (through a lot of repetition) and learned how adjusting different parameters will affect outputs. This growth allowed us to quickly create many images for both inspiration and implementation.
SELECTION OF TOOLS
Adobe Firefly
Adobe firefly was one of the first options we used. But the incorporated AI tool in Photoshop turned to be more useful. Adobe Firefly lacks a proper creativity strength feature which tells the AI how much it can change the original image. The firefly web app was not giving us very good results despite trying many different prompt structures. It had trouble understanding that we wanted something overlayed on top of the image we were feeding it. This meant it was giving results where it was changing the area in the photo to look unrecognizable when we goal was to have the space still be understood as a place on campus.
MIDJOURNEY AI
Having a strong AI image generation tool has been extremely helpful in creating images. The style of images we aimed to make would take days to complete without AI tools assisting us. Midjourney has been the main tool for combining creating images where environment appear to be breaking through the school. Some images are perfect, while others need some human editing to refine. Many of the natural glitches from AI ar embraced because they work with the glitch theme.

PROMPTING
With MidJourney we learned to experiment with prompting. Your prompt has to be as clear as possible avoiding language that is confusing and keeping in mind that everything that you provide has to formatted correctly.
Providing parameters or conditions modify the output file in various ways. Some of these include picture quality, aspect ratio, ‘style’ to mention an specific style to follow, , ,and ‘keep’ to keep an specific thing from the image. The parameters go in the end of your prompt and are separated by commas.