| 摘要 |
A picture coder divides a frame of coded picture data into a critical segment including information needed for decoding other segments in the frame, and at least one non-critical segment including only information needed for its own decoding. The critical segment is made smaller than the non-critical segments, by controlling the segment size directly, for example, or by controlling a quantization step size or a forced intra-coding rate. The small size reduces the probability of the loss of critical information during packet transmission of the coded data, when each segment is transmitted as a separate packet. Alternatively, the size of all segments is reduced when conditions on the communication path indicate a high probability of transmission errors.
|
