Re: Document AI- Key value pairs

[Upkar Singh]

Hello,

I believe what you're looking for to get key-value pairs is this [1] and for tables [2]. There shouldn't be any major differences between Document AI and Google Vision OCR besides that Document AI is intended to be used with other GCP products [3]. I also believe that you can use Google Vision locally as mentioned in the public documentation [4].

Hope this was helpful.

[1]https://cloud.google.com/document-ai/docs/handle-response

[2]https://cloud.google.com/document-ai/docs/process-tables

[3]https://cloud.google.com/document-ai#section-10

[4]https://cloud.google.com/vision/docs/ocr#detect_text_in_a_local_image

On Thursday, October 21, 2021 at 3:23:30 PM UTC-4 sina.bu...@gmail.com wrote:

In the Try section of the Google Document AI, it extracts and shows Key-value pairs but I couldn't find key-value pairs labels in the JSON file.

How can I get the key-value pairs fields or table date file? If it only provides JSON files what is the difference between Document AI JSON file and Google Vision OCR?

Thanks,