Efficient coding strategy for HEVC performance improvement by exploiting motion features

The striking feature of High Efficiency Video Coding(HEVC) Standard is emphasized by 50% bit-rate reduction compared to its predecessor H.264/AVC while keeping the same perceptual imagequality. The time complexity- a congenital issue of HEVC has alsoincreased to intensify the compression ratio. However, it is really ademanding task for the researchers to reduce the encoding time whilepreserving expected quality of the video sequences. Our contributionis to trim down the computational time by efficient selection ofappropriate block-partitioning modes in HEVC using motion featuresbased on phase correlation. In this paper, we use phase correlationbetween current and reference blocks to extract three motion featuresand combine them to determine binary motion pattern of the currentblock. The motion pattern is then matched against a codebook ofpredefined pattern templates to determine a subset of the inter-modes.Only the selected modes are exhaustively motion estimated andcompensated for a coding unit. The experimental outcomesdemonstrate that the average computational time can be down scaledby 30% of the HEVC while providing improved rate-distortionperformance.