I want to Extract the images and its caption from the pdf and save those images with caption in blob storage . Can you suggest me the way to do it . Currently i am thinking of the Object detection model which i have trained on the 20-30 pdf from list of pdf. Will pass the converted image of each page of the pdf to the model and will get the bounding boxes. but don’t whether this will work or not.

Below is my sample screenshot of pdf images and there is no hard rule that all the images will be on the particular page it can vary also.

I also want to extract the caption in the images and save the image with this caption only.
Please guide me. Note : I have subscription for the vertex AI so if any advantage i can use from it let me know .