If you're a noob like me, these web-based tools can help you post-process any pictures you get from text-to-image AI models. Oh, and they don't cost anything!
Yes that's right. The site works, but very slowly. The first picture is recognized for 60 seconds, the second picture for 150 seconds, the third picture for more than 500 seconds.
I decided to install Interrogate CLIP stable diffusion, but this solution does not work for AMD, I need to look for crutches. )
Ah. Gotcha. Yeah, Hugging Face demos are notoriously slow (you usually end up in a queue before your request is processed). So if you're looking to bulk reverse-engineer text prompts, I can see how that's quite a hurdle. If you ever find a better alternative, I'd love to hear about it!
Sure, but it depends on the purpose. If you're looking to reverse-engineer a text prompt for an image, you've got CLIP Interrogator and the like. I wrote about these in the past:
Which site/tool are you referring to exactly? The image-to-text tools I've linked to are cloud-based services and don't depend on the graphics cards you have. You upload an image and they run the analysis and return the results directly on the site. Or are you referring to something else?
Yes that's right. The site works, but very slowly. The first picture is recognized for 60 seconds, the second picture for 150 seconds, the third picture for more than 500 seconds.
I decided to install Interrogate CLIP stable diffusion, but this solution does not work for AMD, I need to look for crutches. )
Ah. Gotcha. Yeah, Hugging Face demos are notoriously slow (you usually end up in a queue before your request is processed). So if you're looking to bulk reverse-engineer text prompts, I can see how that's quite a hurdle. If you ever find a better alternative, I'd love to hear about it!
I find it! ))) https://www.youtube.com/watch?v=2EV5SZ1Klro
Work fast!
Thank you! ))
That's perfect, thanks for sharing the alternative option!
Alternative is Interrogate CLIP for Stable Diffusion, but its only for Nvidia. (
Is there a service that generates an image description from an image?
Sure, but it depends on the purpose. If you're looking to reverse-engineer a text prompt for an image, you've got CLIP Interrogator and the like. I wrote about these in the past:
https://www.whytryai.com/p/clip-interrogator
https://www.whytryai.com/p/image-prompt-sites
If you're looking to generate something like alt tags, something like this should work: https://alttextmagic.com/
Unfortunately, this only works with Nvidia graphics cards. Doesn't work on AMD.
Which site/tool are you referring to exactly? The image-to-text tools I've linked to are cloud-based services and don't depend on the graphics cards you have. You upload an image and they run the analysis and return the results directly on the site. Or are you referring to something else?
O, greate! Thank you.