Title :
Tensor video coding
Author :
Mahfoodh, Abo Talib ; Radha, Hayder
Author_Institution :
Dept. of Electr. & Comput. Eng., Michigan State Univ., East Lansing, MI, USA
Abstract :
In this paper, we exploit the intrinsic tensor-nature of video and propose a Tensor Video Coding (TVC) framework that is based on tensor decomposition. We develop a Progressive Canonical-decomposition Parallel-factor (PCP) framework that is tailored for video representation and coding. Our simulation results show that the proposed TVC approach outperforms state-of-the-art video coding methods that do not rely on motion estimation or compensation.
Keywords :
tensors; video coding; intrinsic tensor-nature; motion compensation; motion estimation; progressive canonical-decomposition parallel-factor framework; tensor decomposition; tensor video coding; video representation; Abstracts; Encoding; Erbium; IEC standards; Tensile stress; Three-dimensional displays; Video coding; CP Tensor Decomposition; Rank-one Tensors; Tensor Decomposition; Video Coding;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on
Conference_Location :
Vancouver, BC
DOI :
10.1109/ICASSP.2013.6637947
