关键词: MPEG-2, H.264, 编码模式, 机器学习, 决策树

Abstract: This paper presents a macroblock（MB） mode decision algorithm for interframe prediction based on machine learning techniques to be used as part of a very low complexity video transcoder.The proposed approach is based on the hypothesis that MB coding mode decisions in H.264 video have a correlation with the distribution of the motion compensated residual in the MPEG-2 video.Authors save the MB information including MB coding modes，CBPC，the mean and variance of MB residual when decoding MPEG-2 video，then save the MB coding modes from H.264 encoder of YUV image.Authors also use data mining tools to exploit the correlation and derive decision trees to classify the incoming decoded MBs into one of the several coding modes in H.264.The proposed approach reduces the H.264 MB mode computation process into a decision tree lookup with very low complexity.

Key words: MPEG-2, H.264, coding mode, machine learning, decision tree