Fast algorithm for the High Efficiency Video Coding (HEVC) encoder using texture analysis

High Efficiency Video Coding (HEVC) is the newest video coding standard for improvement in video data compression. However, the important challenge of the encoding complexity of HEVC is addressed in a new way. For a natural video sequence, a large number of coding blocks are generally skipped in nature. These need not be transmitted and can be generated in the decoder side using reference pictures. An early skip detection technique for HEVC is proposed based on identification of motionless and homogeneous texture-based regions in a video sequence. Entropy is considered as a feature for analysis of coding unit (CU) texture. An entropy difference between coding units is calculated for identification of motionless and homogeneous regions in a video sequence. Complex texture-based regions are generally encoded as asymmetric motion partitioning (AMP) processing units. An early non-AMP detection technique is also proposed for non-skipped coding blocks using the entropy value of a coding unit. The proposed technique achieved a 40% encoding time reduction, compared with the conventional HEVC encoder, with a negligible degradation in video quality.