About Figure 4 in the paper. Illustration of 6 types of interactions in interactive text-to-image workflow

Zeqiang-Lai / Mini-DALLE3

Mini-DALLE3: Interactive Text to Image by Prompting Large Language Models

https://minidalle3.github.io/

298 stars 28 forks source link

About Figure 4 in the paper. Illustration of 6 types of interactions in interactive text-to-image workflow #3

Closed XuRui314 closed 10 months ago

XuRui314 commented 10 months ago

Are all of these generated by mini-DALLE3? I wonder how the 3. Selecting is done. Because in my opinion, every time mini-DALLE3 just generates a new image according to the prompted input, so it‘s kind of impossible for the 2 images to be the same. thx

Zeqiang-Lai commented 10 months ago

All you need to do is adding a selecting tool. See our v3 prompt

XuRui314 commented 10 months ago

I see, this ability is not implemented yet. thx