Forum Stats

  • 3,827,149 Users
  • 2,260,746 Discussions


How do i train custom model for key-value extraction in OCI vision Document AI

Mohamed Rashiq P P-Oracle
Mohamed Rashiq P P-Oracle Posts: 1 Employee
edited Apr 5, 2022 9:37AM in Machine Learning

I'm trying to extract key value pairs from passport using oci vision documentAI service. default model is not extracting key value pairs. it seems there is an option train custom model. Though, I'm not able to see how it can be done for Key-value extraction.

I'm looking for extracting key value pairs from passport, like

firstName: some_firstName

lastName: some_lasttName.

If there is a possiblity of transfer learning, i'm open to that as well

Note: I need to achieve this using oci-vision, not any open-source service