Virtual Background Processor
Interface for virtual background processing.
Platform implementations use ML Kit Selfie Segmentation to:
Capture video frames from the camera
Segment the person from the background
Composite the person with a virtual background
Output processed frames to a new video track
Inheritors
Properties
Functions
Update the frame callback for preview while processing continues. This allows registering a preview callback when the processor is already running.
Start processing the input video stream with the given background.
Start processing with a WebRtcDevice to create output stream. This method creates a virtual video source that can be used with mediasoup to send processed frames to remote participants.
Stop processing and release resources.
Update the virtual background while processing continues.