Motion vector recovery with Gaussian process regression

Asheri, H ; Sharif University of Technology | 2011

592 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/ICASSP.2011.5946563
  3. Publisher: 2011
  4. Abstract:
  5. In this paper, we propose a Gaussian Process Regression (GPR) framework for concealment of corrupted motion vectors in predictive video coding of packet video systems. The problem of estimating the lost motion vectors is modelled as a kernel construction problem in a Bayesian framework. First, to describe the similarity between the neighboring motion vectors, a kernel function is defined. Then the parameters of the kernel function is estimated as the coefficients of a linear Bayesian estimator. The experimental results verify the superiority of the proposed algorithm over the conventional and state of the art motion vector concealment methods. Moreover, noticeable improvements on both objective and subjective measures, on videos with heavy packet loss rates have been achieved
  6. Keywords:
  7. Bayesian estimations ; Bayesian frameworks ; Error concealment ; Gaussian process regression ; Kernel construction ; Kernel function ; Motion vector recovery ; Motion Vectors ; Packet loss rates ; Packet video ; State of the art ; Algorithms ; Bayesian networks ; Estimation ; Gaussian noise (electronic) ; Image coding ; Regression analysis ; Signal processing ; Speech communication ; Vectors ; Gaussian distribution
  8. Source: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 22 May 2011 through 27 May 2011 ; May , 2011 , Pages 953-956 ; 15206149 (ISSN) ; 9781457705397 (ISBN)
  9. URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5946563