For a personal project of mine, I need a tool to describe the contents of an image and generate a list with the data extracted from the image.

For example, the image could be a "digital self" of a supermarket, with n different products. Every product with an image, price and name. Ideally, the AI tool would extract a csv from the image containing the relevant data of each product.

I would need to do that with hundreds of images and thousands of products every day. I plan to do it with Python.

What would be an efficient way to implement this?

I have tried to use google bard and it seems to work just OK. However there is not an official API to implement in python. I also think I shouldn't need a LLM to do it.

I have also tried to use tools that simply extract text from an image. While they work, the position of relevant information for each product could not always be in the same place. I need a tool that is capable of understanding the image, not just extracting text from it.

  • There are many tools that extract keywords from an image. How important is it that the tool matches the price and name? For instance, the answer below does not do that. If important, would you mind rewording the title?
    – Nicolas Raoul
    Dec 28, 2023 at 8:07

1 Answer 1


One possible way is to use this service: https://imagga.com/

It offer API to call and return list of keywords which characterize what is on the image. I personally use it to generate keywords for my photos in Lightroom.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.