作者: A. Sharifinejad , H. Mehrpour
关键词: Search algorithm 、 Motion estimation 、 Algorithm 、 Block-matching algorithm 、 Redundancy (information theory) 、 Encoder 、 Maxima and minima 、 Computer science 、 Redundancy (engineering) 、 Theoretical computer science 、 Data compression ratio 、 Data compression
摘要: The removal of temporal redundancy for reducing the bit rate videos is most time consuming part video encoders. Our proposed algorithm, TWS, improves encoder's speed by at least 42 times with respect to full search algorithm. Moreover, not only TWS (just like algorithm) maintains MSE its minimum value, but it also retains encoded videos' compression highest. development algorithm based on statistical data motion vectors which are extracted from 5 long sequences. According length vectors, observed that they can be grouped in three distinctive ranges constitutes foundation having windows TWS. unlike other fast algorithms does have problem convergence towards local minima distortion function. has almost same values as FS