What inputs does Extract Image Captions support?
This tool currently accepts PDF, Word, PowerPoint, HTML. After upload, Kitlot passes the file through Docling for parsing and conversion.
Extract Image Captions
Extract picture-related blocks and captions from documents.
ProductivityAPIStatic billingFree
Local DoclingStructured output
Selected file
One file at a time. Keep uploads under 20MB.
Accepted formats: PDF, Word, PowerPoint, HTMLPrimary output: structured JSON
About this tool
Extract picture-related blocks and captions from documents.
ProductivityAPIStatic billingPublic
Tool facts
Execution type: API
Credit cost: Free
Billing unit: 1 document
Supported locales: English, Chinese
Last updated: Apr 14, 2026
Related content
Published articles and notes currently linked to this tool.
No published content is linked to this tool yet.
Similar tools
Frequently asked questions
What does Extract Image Captions return?
The primary output is structured JSON. Returns detected image blocks, visual structure, and related metadata.
How to use this tool
1
Choose a file
Open Extract Image Captions and choose one supported document.
2
Run
Kitlot sends the file through the local Docling conversion flow.
3
Review and export
After processing, review, copy, or download the structured JSON result.