I am using Form Recognizer client library for .NET to extract data from a country's national id card. But it seems that it is unable to get the key-value pairs from the card. I have followed https://learn.microsoft.com/en-us/azure/cognitive-services/form-recognizer/quickstarts/dotnet-sdk tutorial.
I have found that there is only one key-value pair and the value array contains all the labels/texts (both key and value).
Do I need to follow another way to extract information from id card? Do I need to train the model with labels using the labeling tool?
Yes, please try training with labels.