Alright, I got this idea from a user at Arc en Ciel discord, I thought it was funny, so I trained it.
I have a kofiā, that is the only way of supporting me and keep my trainings running now. I also take commissions, so if you are interested on a training, contact me or visit my kofi or fiverr listings!
Is really simple just use the activation tags.
burger costumeyou can try to prompt the content of the burger, but in my experience, it doesn't work, prefer squared images or the burger gets tall and the filling looks weird, like the cover.hot dog costumeIt just works, sometimes it has ketchup or lettuce.fries costumeNot the best dataset, it works from the front, not sure on other angles.
You might need to get the weight to 1.2 if you are doing something complicated, is not terribly overfit.
That's it! Generate, leave reviews, funny pictures, etc. They really help to stay motivated, and they support my work and future trainings! Up to R images can get a pin!
Description
Initial version.
FAQ
Comments (2)
hmm,, a second version might want to use real-life photos of such costumes for back/rear views.
Product example images might help
I did, even cats. I'll probably train for anima, I can do an update.







