AI Solution for Extracting Data from Large Scanned PDFs

[vivek_5511]

Subject: AI Solution for Extracting Data from Large Scanned PDFs

Dear Adobe Team,

I hope you're doing well. I wanted to inquire if Adobe has the capability to extract data from scanned PDFs, especially when the file size is large. If this is a challenge, I’d love to connect and discuss a solution—I’ve already built an AI-powered system using Gemini 1.5 Flash that effectively processes scanned textbooks to generate structured notes and provide accurate Q&A.

Key Features of My Solution:
✅ Processes large scanned PDFs into structured study notes
✅ Extracts accurate answers from the provided material, ensuring syllabus-specific learning
✅ Intelligent resource handling—prioritizing user content while offering external support when necessary

If this aligns with your goals, please feel free to DM me. I’d be happy to share insights and explore potential collaborations.

Looking forward to your response.

Best regards,
Vivek Avdhesh Sharma

[Monika Gause]

Did you check out Acrobat?

https://www.adobe.com/acrobat/generative-ai-pdf.html 

[vivek_5511]

I tried adobe but can't processing scanned pdf it can't extract data from it 

[vivek_5511]

@Monika Gause 
my AI project aims that it take scanned pdf of textbook with unlimited size of page perform extraction on it save into temporary memory and provide solution of combine extracted data and AI knowledge

[Monika Gause]

That goal is exactly what Acrobat is meant to do. What size is your textbook?

[vivek_5511]

no limittation

[vivek_5511]

Currently I am student so i have textbook page range from 200-300 pages

[droopydog500]

I moved this to the Acobat forum from Firefly.

    droopy

[vivek_5511]

@droopydog500 
Right now adobe can't take scanned pdf  my project aim is that it takes scanned pdf perform extraction operation on it extract all data from pdf save into memory and provide solution

Note : Adobe takes normal pdf (document-to-pdf) and my project takes scanned pdf there is many difference, and another side it provide solution using extracted data