Skip to main content

Is it possible to automate metadata extraction from photo and video files?

  • March 27, 2026
  • 1 reply
  • 160 views
thomasdeely Box
Forum|alt.badge.img

Video files aren’t supported yet. For photos, it should work as long as they’re in one of the supported formats: JPEG, JPG, TIF, TIFF, or PNG 

 

This could be along the lines of EXIF data too, which isn't really an AI thing, but rather just a data layer Box could scrape. https://photographylife.com/what-is-exif-data

 

Box Consulting has a custom offering for both one-time EXIF scrape from a set of existing files and go-forward application of EXIF data. 

 

A workaround is extract from the video transcript, although you don’t get the full context of the video. 

    1 reply

    thomasdeely Box
    Forum|alt.badge.img
    • thomasdeely Box
    • Author
    • Sr. Community Manager
    • July 22, 2026

    Can Box Auto-Populate Metadata from Images?

    Yes, absolutely! Box has powerful, native AI-driven capabilities that can automatically extract structured data from images and apply it directly to your metadata templates.

    This is primarily achieved through Box Extract (which utilizes Box AI Standard and Enhanced Extract Agents) and the Autofill Metadata feature in Box Preview:

    1. Supported Image Formats: Box Extract and Box AI Autofill support extracting metadata from several image types, including .png, .tiff, .tif, .jpg, .jpeg, and .webp.
    2. How It Works (OCR & Advanced AI): Box Extract combines advanced Optical Character Recognition (OCR) with state-of-the-art Large Language Models (LLMs). Instead of just pulling raw text, the AI understands the context, layout, and meaning of the image (such as reading an invoice, receipt, or ID card) to accurately map information to specific metadata fields.
    3. Autofill in Box Preview: If enabled by your administrator, you can open an image in Box Preview, open the Metadata tab in the right-hand sidebar, and click Autofill. Box AI will analyze the image and suggest values for your metadata template fields, which you can review and save with a single click.
    4. Automated Folder-Level Extraction: You can configure Custom Extract Agents and attach them to up to 10 source folders. Any supported image or document uploaded to those folders will automatically have its metadata extracted and applied natively.
    5. API & Integrations: For custom applications, the Box AI Extract API allows you to integrate this capability directly into your custom technology stack or Salesforce workflows.

    Note: The ability to create and configure Custom Extract Agents and use Autofill is available on Enterprise Advanced plans, while the underlying APIs are available on Business plans and above.

    Recommended Box Education Training

    To help you develop your skills and get the most out of Box's metadata and AI capabilities, I highly recommend exploring these self-paced courses and resources from Box Education:

    • Configuring and managing metadata
      • Format: Self-Paced Course
      • Estimated Duration: 20 minutes
      • Link: Register Here
      • Description: Learn how to design, configure, and manage metadata templates to structure your business content.
    • Foundations of Box AI
      • Format: Self-Paced Course
      • Estimated Duration: 30 minutes
      • Link: Register Here
      • Description: Understand the architecture of Box AI, including how it handles metadata extraction and multi-document queries.
    • Building custom AI agents with Box
      • Format: Self-Paced Course
      • Estimated Duration: 30 minutes
      • Link: Register Here
      • Description: Learn how to configure and deploy custom AI agents (including extraction agents) using Box AI Studio.

    Additional Resources & Support

     

    • Technical Support: If you encounter any issues or need help setting up metadata templates, you can submit a support ticket at support.box.com.
      Terms of ServiceAccessibility statement