ICR Documentation

Async ICR (Intelligent Content Recognition) API

Overview

The Async ICR (Intelligent Content Recognition) API enables asynchronous analysis and understanding of documents and images using configurable, pipeline-driven processing. Beyond text extraction, the API is designed to identify, interpret, and structure information from a wide variety of inputs, including documents, forms, certificates, and contextual images.

The API supports intelligent processing across:

Structured, semi-structured, and unstructured content
Printed, handwritten, and mixed format documents
Single image and multi-page inputs
Images and documents in multiple languages
Visual and contextual cues beyond text (where applicable)

All processing is performed in the background, allowing clients to submit documents or images without blocking their workflows. Results can be retrieved by polling the request status once processing is complete.

Key Characteristics

Asynchronous by design for non-blocking workflows

Pipeline-driven intelligence, allowing document and use-case-specific logic

Supports documents and images, not limited to text-only inputs

Handles standards, non-standard, and complex formats

Enables bulk uploads

Scales for high-volume and enterprise-grade workflows