Document Understanding Features

Application-ready AI models

Extract text from scanned documents, mobile uploads, PDFs, and more

OCI Document Understanding uses optical character recognition (OCR) and other advanced models to automatically extract text from a variety of document files, including documents that are rotated, tilted, or shaded, to support quality issues often found in expense processing and customer onboarding.

Extract tables and key fields from documents

Automatically identify and extract table structure from documents, including the row and column relationships within your table. For expense and identity documents, OCI Document Understanding can identify and extract key-value pairs from invoices, receipts, passports, driver’s licenses, and health insurance ID cards.

Categorize documents

Identify and classify documents into common categories, such as invoice, receipt, and résumé. Common applications include expense processing and enhanced search and retrieval of documents.

Multilingual support

OCI Document Understanding’s pretrained models for optical character recognition and key-value pairs support multiple languages, including Arabic, Chinese, Dutch, English, French, German, Hebrew, Japanese, Portuguese, Russian, Spanish, and Ukrainian.

Custom-trained models

Create custom models for key-value pairs and document classification. With OCI Document Understanding, customers can train, evaluate, deploy, and analyze models with their own data.

Secure and accessible

Purpose-built for security and privacy

OCI Document Understanding upholds customer privacy with models that don’t store any data for training, debugging, or other purposes.

Easy to deploy

OCI Document Understanding is a versatile service that can be called via REST APIs, multiple SDKs (including Python and Java), or the OCI command line. Developers can easily deploy a scalable document service without having expertise in data science or machine learning.

Dedicated endpoints

Provision dedicated endpoints for greater control and the ability to meet high throughput requirements for OCI Document Understanding workflows.