| 摘要 |
Video coding methods and apparatuses are provided that make use of various models and/or modes to significantly improve coding efficiency especially for high/complex motion sequences. The methods and apparatuses take advantage of the temporal and/or spatial correlations that may exist within portions of the frames, e.g., at the Macroblock level, etc. The methods and apparatuses tend to significantly reduce the amount of data required for encoding motion information while retaining or even improving video image quality.
|
