Polybos Media Engine
AI-powered video metadata extraction
Extractors:
Metadata
Scenes
Faces
Transcript
Objects
CLIP
OCR
Motion
Objects:
YOLO
Qwen
All
None
Transcript Context
(helps Whisper with names, terms, accents)
Scene Context
(helps Qwen identify people & objects)
Re-run
Change File
Select a video file:
Up
Loading...
Run
Processing...
○
Metadata
○
Transcript
○
Scenes
○
Faces
○
Motion
○
Objects
○
CLIP
○
OCR
○
Telemetry
Video
Metadata
Location
Export GPX
Transcript
—
Faces
—
—
Objects
—
Scene Descriptions
Text (OCR)
—
Camera Motion
—
—
Scenes
—
JSON
Error