A New Spatial Downsampling Algorithm for DCT-Domain Video Transcoder

摘要

Spatial resolution reduction of compressed video is highly desirable in video transcoding. Many studies have addressed this problem, but they either cause drift error or have high computation complexity. A new spatial downsampling algorithm for DCT-Domain video transcoder is proposed in this paper, in which L×M sub-blocks of N×N are combined into a L×N rows and M×N columns blocks in DCT-Domain firstly, by extracting the left-top N×N low-frequency coefficients from this block, the new N×N downsampled block is the obtained. Experimental results confirm that the proposed algorithm performs better than other algorithms in terms of PNSR of reconstructed video sequences, furthermore, the computation complexity reduced greatly. sheet, as illustrated by the portions given in this document.