Rate-distortion based video coding mode selection foreseeing the esitmation of bit rate and distortion using a simplified transform on low activity prediction residuals

摘要

In MPEG4-AVC encoding an advanced prediction with multiple prediction modes is used to increase the prediction efficiency. Known rate/distortion optimisations control the coding mode decision by evaluating the cost of each candidate coding mode and selecting the mode that yields the minimum RDO cost. The process for determining the RDO cost requires carrying out a whole encoding process to produce the coded bits and reconstructed pixels needed, requiring high processing power. The inventive rate (R1, R2) and distortion (D1, D2) calculation can be used to determine (BTEST1; BTEST2, S3) efficiently the RDO cost in the frequency domain, in a picture content-dependent manner, using an efficient rate/distortion optimised mode decision that is based on a simplified transform (STR, SITR) to obtain the lower frequency coefficients. Thereby the entropy coding can be omitted.