17 March 2017 Video denoising using low rank tensor decomposition
Author Affiliations +
Proceedings Volume 10341, Ninth International Conference on Machine Vision (ICMV 2016); 103410V (2017) https://doi.org/10.1117/12.2268435
Event: Ninth International Conference on Machine Vision, 2016, Nice, France
Reducing noise in a video sequence is of vital important in many real-world applications. One popular method is block matching collaborative filtering. However, the main drawback of this method is that noise standard deviation for the whole video sequence is known in advance. In this paper, we present a tensor based denoising framework that considers 3D patches instead of 2D patches. By collecting the similar 3D patches non-locally, we employ the low-rank tensor decomposition for collaborative filtering. Since we specify the non-informative prior over the noise precision parameter, the noise variance can be inferred automatically from observed video data. Therefore, our method is more practical, which does not require knowing the noise variance. The experimental on video denoising demonstrates the effectiveness of our proposed method.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Lihua Gui, Lihua Gui, Gaochao Cui, Gaochao Cui, Qibin Zhao, Qibin Zhao, Dongsheng Wang, Dongsheng Wang, Andrzej Cichocki, Andrzej Cichocki, Jianting Cao, Jianting Cao, } "Video denoising using low rank tensor decomposition", Proc. SPIE 10341, Ninth International Conference on Machine Vision (ICMV 2016), 103410V (17 March 2017); doi: 10.1117/12.2268435; https://doi.org/10.1117/12.2268435


Fractal-like video coding with weighted summation
Proceedings of SPIE (February 27 1996)
A fast non-local image denoising algorithm
Proceedings of SPIE (March 03 2008)
Towards a multimedia remote viewer for mobile thin clients
Proceedings of SPIE (February 08 2011)
High-level feature extraction in JPEG compressed domain
Proceedings of SPIE (November 02 2004)

Back to Top