With the video layers allocation header extension, we learn when WebRTC has decided not to send a stream, so we can switch away sooner than through detecting a lack of data. Additionally, target rates from the header let us estimate bandwidth needed to receive the stream from the first packet received. Even without client updates to send the header, allocating in response to changes in properties of the stream allows video forwarding to start sooner, and will reduce delay when switching from receiving layer 2 camera video to layer 1 presentation. |
||
|---|---|---|
| .. | ||
| src | ||
| Cargo.toml | ||