OCR and beyond: Reading data from unstructured sources

[amotl]

Pitch

Is unstructured text input out of scope for omniload? Who defines its features boundaries?

Example: PDF extraction

What about loading a bunch of PDF documents into a spreadsheet, where each row represents a document, and each column cell includes content from a specific section of the document?

References

• https://github.com/baidu/Unlimited-OCR (discuss)

[amotl]

xberg

Document intelligence: text, tables, metadata from 91+ formats with optional OCR.

https://github.com/xberg-io/xberg
https://docs.xberg.io/