"Detect text from picture" in the "Edit media" dialog of the Mastodon PWA is not a bad way to extract code from screenshots. It has some predicable issues such as mistaking digit 1 for letter l and 0 for @, but not too bad for short code segments.

Recommend 8/10

#mastodon #pwa #imageToText #image #text

"Midjourney now offers a clever twist on its #AI image software, using its powerful machine-learning algorithms to generate text descriptions of already-existing images. This makes detailed image captions trivially easy to make and could totally change the game for blind people when it comes to pictures. "

Why #Midjourney’s New #ImageToText Generator Is an #Accessibility Home Run | #AI #GenerativeAI
https://www.lifewire.com/why-midjourneys-new-image-to-text-generator-is-an-accessibility-home-run-7377344

Here we go AGAIN!

Midjourney just released the "/describe" prompt. Upload a photo (#1) and Midjourney does -- imageTOtext -- by "reading" your image and creating four, distinct text prompts based on your original image that Midjourney will use to "create" four whole new image prompts in a 4x4 grid.

You upscale the looks you want to see from the interpretation (#2, #3, #4).

Here's more of what it looks like! This is wild stuff, and already crashed the server!

#AI #Midjourney #ImageTOtext

Here we go!

Midjourney just released the "/describe" prompt.

You upload a photo (#1) and Midjourney then does -- imageTOtext -- by "reading" your image and then creating four, distinct text prompts based on your original image that Midjourney will use to "create" four whole new image prompts in a 4x4 grid.

You then upscale the looks you want to see from the interpretation (#2, #3, #4).

Here's what it looks like! This is amazing, and already crashed the server!

#AI #Midjourney #ImageTOtext

Few corrections from the text scrape that I missed.

READY SPREADY GO -is the one on the far left I'm pretty sure.

Which of course leaves SALT DISNEY making much more sense!

#winter #snow #ice #WordPlay #Puns #Puntastic

#GoogleLens #ImageToText #Whoa #NearlyGotIt #Blimey #Nerdy

#ChatGPT is fun, esp when the output is pared down and fed into DallE or Midjourney etc.

What I'd really like to see though is an #AI tool that can describe an uploaded image in text. Does any such tool exist?

#AIGenerator #imagetotext

I've been experimenting with facilitating collaborations between AI generators by having one tool generate a text description of something, which I can then feed into a text-to-image generator.

What I'd really like to find is an image-to-text generator: upload an image, and have the AI describe the image in text. DOES THIS EXIST?

#AI #ImageToText #DallEButBackwards

Firefox's ability to extract and copy text from an image is rather impressive. Because of all the different fonts & typefaces used. This one presented a challenge.

#Firefox #AI #ImageToText