I am uploading a file and extracting data from it to be used later.
Now, images are not returned.
Any idea how I could still also extract images?
Example:
PDF that is a course book on AI.
I want to not only extract text but also figures. Is it possible that Gemini returns images too? Maybe through urls?
How would you go about this challenge?
Hi @robo,
Sorry for the delay in response. The Gemini File API does not support image extraction from PDFs Instead you can programmatically extract images using a Python library like pypdf and write your prompt in such a way that your text will point out the image with the reference (e.g., Fig 1.7) to get information about them (e.g., descriptions, object recognition, text extraction from the image).
Thank you!