- Aug 13, 2019
- 797
- 6,748
Yes, indeed but the only downside for now is th elack onf consistency between renders,specially the faces. If you can tell your bot to always make the same face and body features it'll be possible to meI was using Image Guidance with a high strength to create similar pictures, the prompt is almost irrelevant in that case.
But today I found out that the Universal Upscaler gives even better results because the images are more realistic and closer to the originals.
Looks like any game developer can now use those tools to increase the quality of their products. I even started to think about making a game on my own. It's getting easier and easier every day.
Here are some samples:
View attachment 3370085
View attachment 3370087
View attachment 3370090
View attachment 3370091
View attachment 3370092
- May 6, 2017
- 26
- 262
I think there's a way to achieve that by training your own model and then using it to create new content with prompts. For training you would just need lot of pictures to learn AI about the details of the face.
Finally feeling a bit more motivated to start working on art work again.
You don't have permission to view the spoiler content.
Log in or register now.
- Jan 2, 2019
- 7
- 40
- Jul 31, 2017
- 1,063
- 20,506
- Mar 16, 2018
- 3,933
- 28,269
- Jun 5, 2017
- 93
- 1,197
