2014/04/12 by Yehuda Dar, Alfred M. Bruckstein, Alfred M. Bruckstein⋆ +2
Computer Science · Mathematics · #Advanced Image Processing Techniques #Advanced Vision and Imaging #Algorithm #Artificial intelligence #Autocorrelation #Block-matching algorithm #Coding (social sciences) #Coding gain #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Data compression #Decoding methods #FOS: Computer and information sciences #Frame (networking) #Inter frame #Mathematics #Motion compensation #Motion interpolation #Multimedia (cs.MM) #Reference frame #Statistics #Telecommunications #Video Coding and Compression Technologies #Video processing #cs.CV #cs.MM
paper · pdf · doi:10.48550/arxiv.1404.3290
arxiv created 2014/04/12 · openalex publication_date 2014/04/12 · arxiv updated 2014/04/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Block-based motion estimation (ME) and compensation (MC) techniques are\nwidely used in modern video processing algorithms and compression systems. The\ngreat variety of video applications and devices results in numerous compression\nspecifications. Specifically, there is a diversity of frame-rates and\nbit-rates. In this paper, we study the effect of frame-rate and compression\nbit-rate on block-based ME and MC as commonly utilized in inter-frame coding\nand frame-rate up conversion (FRUC). This joint examination yields a\ncomprehensive foundation for comparing MC procedures in coding and FRUC. First,\nthe video signal is modeled as a noisy translational motion of an image. Then,\nwe theoretically model the motion-compensated prediction of an available and\nabsent frames as in coding and FRUC applications, respectively. The theoretic\nMC-prediction error is further analyzed and its autocorrelation function is\ncalculated for coding and FRUC applications. We show a linear relation between\nthe variance of the MC-prediction error and temporal-distance. While the\naffecting distance in MC-coding is between the predicted and reference frames,\nMC-FRUC is affected by the distance between the available frames used for the\ninterpolation. Moreover, the dependency in temporal-distance implies an inverse\neffect of the frame-rate. FRUC performance analysis considers the prediction\nerror variance, since it equals to the mean-squared-error of the interpolation.\nHowever, MC-coding analysis requires the entire autocorrelation function of the\nerror; hence, analytic simplicity is beneficial. Therefore, we propose two\nconstructions of a separable autocorrelation function for prediction error in\nMC-coding. We conclude by comparing our estimations with experimental results.\n