CLARIFICATION OF AI WORKFLOW (Patreon)

If someone thinking that this is just a button "Make it Beautiful" than this is wrong, because when i said that process sometimes takes even more time than jsut regular render, that was not the joke, not embellishment, to add contrast to sentences, it's just fact. This is how it works for me.

Now step by step.
1 image: This is how you starting. In blender addon uploading core/source image, setting up parameters of AI, like amount of "creativity"(range of generation). Like parameter responsible for how much the prompt affects the image. Usually first 10-15 images i just trying to find balance between Acceptable and Fucking CRINGE. Most of the time it's cringe) Generation of image taking alot of time. Like in not higher quality preset it takes usually from 1 to 3 minutes per frame.

2-4 images: Just take a look on Cassie's processing. As i said most of the time AI generating cringe. And in process of generating i look in to image and marking places that i will cut off later(I just combine small amount of images to get you all understanding how exactly this is working)
After i geenrated like 40-50 images i start building it up in photoshop with maks on opiginal image, from one i taking forehead part, from another one i cutting of balls, from third i taking neck sample and etc. So basicaly i do not do it how most of the people think, like you uploaded image, pressed the button - getting ready image to post, and here's fucking no! :D you actually doing some kind of layering of generated fragments and mixing it up with source image. Like it was before deepfaces with celebrety pictures, when people attached someones face to different body, so this is the same thing, but more complicated. There is still no that wonder button "Make it Awesome" You still need to put alot of time and effort in to it to get best result.

Conclusion: I do not deny that many who use AI do it lazily, do not remove generation errors, do not bother at all, they doing generations based on the work of other people and pass it off as their own. I trying to do everything right. In fact, the AI will not do the basic work for me. Since in order for the AI to understand what I want from it, I still need to make the initial sample, base render in the best possible quality, the better and cleaner images, the easier it will be for the AI to understand what is shown there in the picture and what needs to be done and process it.

Therefore, I have full control over the process, and everything that you see in the final image was selected and finalized by me. I hope there will be no more questions and baseless nit-picking about this. But if you have any other questions, or clarifications, then ask them in the comments.