Extract information accurately from documents using computer vision
summary
Amazon Textract extracts printed text, handwriting, tables and key‑value pairs and adds two new APIs (Analyze Expense, Analyze ID) plus a natural‑language Queries feature to return question/answer pairs. Customers like Black Knight automate mortgage docs and Anthem processes thousands of claims per day achieving ~80% automation (aiming for >90%). Demo builds a CloudFormation/Step Functions pipeline that runs Textract + Comprehend, processes 170 uploads and outputs 5,800+ extracted data points into a database within minutes, reliably pulling values like pay‑stub gross pay and borrower SSNs without custom templates.
AI-written summary · about
chapters (6)
- 0:00Introduction to Textract
- 1:16Value and Capabilities
- 4:40Customer use cases
- 7:33Deep dive into features
- 19:10Live demonstration
- 41:38Getting started resources
related talks
01h 00mExtract information using the Queries feature in Amazon Textract
32mAutomate data extraction & analysis with intelligent document processing
35mAutomate your document pipeline with intelligent document processing
30mIntelligent document processing: Extract data & insights from documents
26mEmpower your business use cases
32mAutomate data extraction & analysis with intelligent document processing
30mIntelligent document processing for the insurance industry
41mInsights into patient health with