Having my books Reconfigure and Cont3xt I am always using them and the story I created as a way to explore other mediums. I started drawing a few years ago to see if I could maybe go manga with it, that included buying some action figure mini drawing mannequins that have some tech and weapon props. I created some meta humans and have a unity based mocap suit from a kickstarter to see if I could produce anime. Audio voice generation has been another thing I tried. Each of these projects would be a significant amount of time, effort and probably cost too, but delving in as a mini proof of concept gives me an appreciation of what would be needed.
Midjourney and the AI image generators have offered another option to explore, an very quickly too. One of the challenges is to get the same character to appear in each generated image (i.e. story board). I decided to sort that out later in this POC. Though I did suggest that the actress to play Roisin should be Ruby Rose to get a potential consistent likeness.
Having got this image, which was pretty good for a first attempt, with very little text in the prompt I popped over to runway.ml and tried its new gen-2 image to video generation. This time with NO PROMPT at at all! It decided she should look up and stare into the distance, which in the book she is doing as she is being chased and very aware of her surroundings as she tries to outwit those after the amazing tech she has discovered. They only generate short clips on the free accounts, but those few seconds I found to be quite astounding and bewitching.
GenAI is obviously on a major drive into all sorts of areas, but the ability to potentially use tools like this to create a film version of my books on my own time and budget is emerging as a strong contender, unless Netflix or Prime decide they would like to do it instead 🙂
Leave a Reply