Title :
Compute-resource allocation for motion estimation in real-time video compression
Author :
Yeh, Joseph ; Wawrzynek, John
Author_Institution :
Dept. of EECS, California Univ., Berkeley, CA, USA
Abstract :
We present a novel method for allocating computational resources in motion estimation for real time video compression that attempts to minimize prediction error over the whole image. The method makes decisions about how much compute power is allocated to motion estimation for each particular block based upon past observations of the input and output data of each task. We present results that demonstrate the ability of our method to improve the resulting quality of encoded video under a real-time computational constraint.
Keywords :
data compression; motion estimation; prediction theory; resource allocation; video coding; compute-resource allocation; motion estimation; prediction error minimisation; real-time computational constraint; real-time video compression; video quality; Compression algorithms; Computer architecture; Concurrent computing; Delay effects; Image coding; Motion estimation; Multimedia computing; Parallel architectures; Resource management; Video compression;
Conference_Titel :
Signals, Systems and Computers, 2004. Conference Record of the Thirty-Seventh Asilomar Conference on
Print_ISBN :
0-7803-8104-1
DOI :
10.1109/ACSSC.2003.1292247
