Intofantasy
Senior Member
- Jan 18, 2013
- 991
- 334
Is that the original image?Hi!
So basically I have a question, let's say I have a picture. Example below.
Is it possible to upload the picture to a AI software, that gives me the prompt exactly in this picture. So I can reprompt it with the different designs?
Does anyone have any knowledge about this much appreciated.
View attachment 294169
Where? Please give the instructions.yes, upload it and ask for the prompt . reverse engine technique.
Thank you mate, will check this one.Is that the original image?
Several stable diffusion frontends like Dreamfusion and Automattic111 embed the prompt in the png metadata of the image itself.
So checking that should be a good place to start.
Otherwise you can use a CLIP-Interrogator tool. Here's one:
https://huggingface.co/spaces/pharmapsychotic/CLIP-Interrogator
Could you please give me the instructions I'll do it immediately, much appreciated! I'm just trying to figure this whole thing out.Yes, you can use Midjourney to describe the image and get a prompt. Then just use the same prompt to generate an image.
If you are in the discord server you can use the describe command to upload an image. It will give you 4 prompts based on that.Could you please give me the instructions I'll do it immediately, much appreciated! I'm just trying to figure this whole thing out.
Thank you for this. I had no idea Midjourney could do that.If you are in the discord server you can use the describe command to upload an image. It will give you 4 prompts based on that.
https://docs.midjourney.com/docs/describe