DocumentCode :
2492788
Title :
Skin region based method for reducing motion estimation time
Author :
Shirali-Shahreza, M. Hassan ; Shirali-Shahreza, Sajad
Author_Institution :
Virtual Educ. Grad. Coll., Amirkabir Univ. of Technol., Tehran, Iran
fYear :
2010
fDate :
15-18 Dec. 2010
Firstpage :
254
Lastpage :
259
Abstract :
Video compression algorithms are time consuming processes. A key part of video compression algorithms which is also time consuming is motion estimation (ME) and computation of motion vectors (MVs). During this process, a series of candidate blocks should be searched to find the best motion vectors. In this paper, we propose a method for reducing the motion estimation process time. The idea of our method is to compare skin regions of candidate blocks with current frame block and skipping some candidate blocks during the search. We propose some conditions for comparing the skin regions of two blocks and describe how we can check these conditions quickly. Our method provides two parameters that can be used to create a balance between the processing time and the motion estimation accuracy.
Keywords :
motion estimation; object detection; skin; video coding; motion estimation; motion vectors; skin region; video compression; Image color analysis; Motion estimation; Motion measurement; PSNR; Pixel; Skin; Video compression; Motion Estimation; Skin Detection; Video Compression;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing and Information Technology (ISSPIT), 2010 IEEE International Symposium on
Conference_Location :
Luxor
Print_ISBN :
978-1-4244-9992-2
Type :
conf
DOI :
10.1109/ISSPIT.2010.5711790
Filename :
5711790
Link To Document :
بازگشت