Title :
An improved DVC scheme based on motion activity classification
Author :
Ye, Feng ; Men, Aidong ; Zhang, Xinming ; Wen, Ge ; Di, Jinhong
Author_Institution :
Sch. of Inf. & Commun. Eng., Beijing Univ. of Posts & Telecommun., Beijing, China
Abstract :
In this paper, an improved DVC scheme that uses backward channel aware motion estimation is presented. A novel key frame coding method and a macroblock classification method are proposed, which are based on motion activity classification. Then, these methods then are used for mode-decision of switch motion estimation method at the decoder. We further present a 3DRS motion estimation based on motion activity classification to improve the accuracy of reconstructed key frames and reduce the complexity of our DVC scheme. The experimental results confirm our analyzing results and show that our coding scheme can achieve significant coding efficiency gain compared with state-of-the-art distributed video codec while still keeping low encoding complexity.
Keywords :
channel estimation; image classification; image reconstruction; motion estimation; video coding; 3DRS motion estimation; DVC scheme; backward channel aware motion estimation; coding efficiency gain; distributed video codec; key frame coding method; key frames reconstruction; low encoding complexity; macroblock classification method; mode-decision; motion activity classification; switch motion estimation method; Computational complexity; Decoding; Encoding; Motion estimation; Transforms; Video coding; 3DRS motion estimation; Distributed Video Coding (DVC); motion activity classification;
Conference_Titel :
Network Infrastructure and Digital Content, 2010 2nd IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-6851-5
DOI :
10.1109/ICNIDC.2010.5657974