We're so close to making that a possibility! Just need to be able to generate multiple consistent characters in the same image, and multi lip sync and ... a bunch of other things, maybe in another year lol
This was made with image2image, the people on the left are sposed to be a metal band(all guys), the girl on the right was in the image and the chihuahua in a tutu, I forget the prompt, but I didn't say anything about Chihuahua headed girl lol and it turned the lead singer into a girl too lol
Comments