1.5 Prerequisites/Preconditions

  • Article

This protocol specifies only the payload format for RTVideo video streams. This protocol requires the establishment of an RTP stream, a mechanism to obtain RTVideo video frames for it to packetize, and a mechanism to render RTVideo video frames that it has depacketized.

Higher layers are required to provide RTVideo frames with the following added information about each frame:

I-frame Flag: Specifies whether the frame is an I-frame.

SP-frame Flag: Specifies whether the frame is a Super P-frame (SP-frame).

cached frame Flag: Specifies whether the frame is a cached frame.

sequence header: This is required for each I-frame. It is not needed for other frame types.

Higher layers are required to provide video frames in referencing order. Frames being referenced are required to be provided earlier than frames referring to them.

Higher layers are also required to respect the following assumptions:

  • An I-frame does not have any reference frame.

  • I-frames and SP-frames are cached frames as well.

  • An SP-frame refers to the previous cached frame.

  • A P-frame refers to the previous P-frame, SP-frame, or I-frame.

  • A B-frame refers to the previous P-frame, SP-frame, or I-frame.