Ultra-low latency video streaming

A method, implemented by a computing device, for low-latency streaming of media content using a lossless network protocol, the method comprising:

• streaming a media stream, comprising encoded video data, to a plurality of streaming clients via the lossless network protocol, wherein the lossless network protocol provides reliable delivery of the encoded video data to the plurality of streaming clients;
• determining whether any of the plurality of streaming clients has fallen behind in streaming the media stream;
• and
• upon determining that a streaming client has fallen behind:
• selectively dropping a portion of video data to be streamed to the streaming client, wherein the selectively dropping the portion of the video data is performed based on both scalability information of the video data and long term reference (LTR) frame information of the video data;
• wherein the determining whether any of the plurality of streaming clients has fallen behind the media stream is performed without using per-client quality feedback from the plurality of streaming clients.

A method, implemented by a computing device, for low-latency streaming of media content using a semi-lossy network protocol, the method comprising:

• streaming a media stream, comprising encoded video data, to a plurality of streaming clients via the semi-lossy network protocol, wherein the semi-lossy network protocol uses a plurality of delivery modes for corresponding different types of encoded video data, wherein each of the plurality of delivery modes provides a different level of reliability;
• wherein the streaming comprises:
• transmitting a first type of encoded video data to the plurality of streaming clients using a first delivery mode, wherein the first delivery mode uses a first number of retries;
• and
• transmitting a second type of encoded video data to the plurality of streaming clients using a second delivery mode, wherein the second delivery mode uses a second number of retries different from the first number of retries;
• wherein the streaming is performed without using per-client quality feedback from the plurality of streaming clients.