OCR-powered metadata extraction with Box AI and MCP

Box AI has introduced OCR support for image files in its structured metadata extraction endpoints, enhancing the capabilities of its API by allowing direct processing of TIFF, PNG, and JPEG files without the need for conversion to PDF. This update supports multiple languages, including English, Japanese, Chinese, Korean, and Cyrillic scripts, and is available in both standard and enhanced structured extraction endpoints, although it is not part of the freeform extraction API. This development streamlines the workflow for developers by eliminating a conversion step, enabling the direct extraction of structured metadata from image-based documents like receipts, invoices, and forms. Box AI can now suggest metadata schemas for image files, facilitating the creation of templates that can be used to extract data with improved accuracy, especially when using sophisticated models for complex layouts. The integration also supports multilingual document processing, broadening the applicability of Box AI's document management systems across various markets.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.