Midjourney is an independent research lab that produces a proprietary artificial intelligence program under the same name that creates images from textual descriptions, similar to OpenAI’s DALL-E and Stable Diffusion.
Here is my journey using MidJourney to generate digital art from text prompts. In my image example below I asked MidJourney to generate an image based on lyrics from the song Fireflies by Owl City.
“a thousand hugs From ten thousand lightning bugs, As they tried to teach me how to dance, A foxtrot above my head, A sock hop beneath my bed, A disco ball is just hanging by a thread.”
It will then generate 4 sample images. From there you can choose to re-do the generation of the 4 sample images or choose an image to upscale into a single image.
After you have found an image you like you can choose to upscale the image making it larger and more detailed. I liked the first image in the top left so I chose that to upscale which you can see below.
You can see from the image above after upscaling MidJourney will add more details and change it slightly. From here you can keep the image as is or create more variations.
Below is a gallery of some of my favorite generations I’ve done so far.