Detect and extract visual elements (tables, figures, charts, diagrams) from PDF pages. Returns bounding boxes and classifications for each detected element. Optionally includes symbol-level detection for technical documents.
API key must be provided in the Authorization header
URL of the PDF to detect elements from
Source ID of an already-indexed document
Page range to process (e.g. "1-10", "5,8,12")
Include symbol-level detection for technical documents
Regex pattern to filter detected element types
Successful Response
Detection job ID
Job status: queued, processing, completed, failed
Processing progress percentage (0-100)
Detection results with bounding boxes and classifications