Agree
Look into enabling the "useEncodedMedia", this way Kurento will act
as a media relay, i.e. the video out of the camera will be sent
as-is to the receivers. Otherwise (if you don't enable
"useEncodedMedia"), Kurento would be re-encoding all video and
audio, which causes a lot more CPU usage.
If you can control the codec and other settings in the camera, it's
best to enable useEncodedMedia in the PlayerEndpoint, and control
all encoding parameters directly from within the camera:
let player = await
pipeline.create('PlayerEndpoint', {
uri: 'rtsp://localhost:5000/video',
useEncodedMedia: true,
});
Note that, specifically for WebRTC, the best is to use H.264 with
Contrained Baseline profile, level 3.1: