Correlation Noise Modeling for Efficient Pixel and Transform Domain Wyner–Ziv Video Coding
Citations
110 citations
Cites background from "Correlation Noise Modeling for Effi..."
...The Laplacian parameters are estimated online at the coefficient level (see [21] for more details)....
[...]
87 citations
Cites methods from "Correlation Noise Modeling for Effi..."
...In this step, full search motion estimation with modified matching criteria is performed [3]; the criteria include a regularized term that favors motion vectors that are closer to the origin....
[...]
...The frame interpolation module generates the side information YWZ, an estimate of the XWZ frame, based on two references, one temporally in the past (XB) and another in the future (XF), as follows: 1....
[...]
...The residual statistics between correspondent coefficients in XDCTWZ and YDCTWZ is assumed to be modeled by a Laplacian distribution; the Laplacian parameter is estimated online at the decoder, for each DCT coefficient, based on the residual between the two motion compensated reference frames used…...
[...]
80 citations
Cites methods from "Correlation Noise Modeling for Effi..."
...In [31], a method is proposed for online estimation at the decoder of the correlation model....
[...]
66 citations
43 citations
References
4,165 citations
3,288 citations
"Correlation Noise Modeling for Effi..." refers background or methods in this paper
...The extension of Slepian–Wolf coding for lossy compression is well known as Wyner–Ziv (WZ) coding [3], which deals with the lossy source coding of when some side information is available only at the decoder....
[...]
...In [3], Wyner and Ziv show that there is no increase in the transmission rate if the statistical dependency between and is only explored at the decoder compared with the case where it is explored both at the decoder and the encoder, notably, if and are jointly Gaussian and a mean-square error distortion measure is considered....
[...]
1,342 citations
"Correlation Noise Modeling for Effi..." refers background or methods in this paper
...For example, in [4], [7], and [8], the CNM parameters for the transform-domain WZ (TDWZ) video coding scheme are obtained through a training stage using several video sequences (one parameter per transform band is computed)....
[...]
...A similar approach to the one mentioned [4] is...
[...]
...In [4], the CNM parameter for a PDWZ video coding scheme is said to be estimated at the decoder based on previously decoded frames statistics; however, it is not specified in [4] how this task is precisely performed....
[...]
...1 matches with the PDWZ video codec architecture in [4]....
[...]
...Knowing the CNM and its parameters at the encoder allows to perform, for example, rate control at the encoder; however, in a WZ video coding architecture with decoder rate control, as the architectures in [4] and the one used in this paper, this is not a relevant feature....
[...]
1,280 citations
"Correlation Noise Modeling for Effi..." refers methods in this paper
...From the Information Theory, the Slepian–Wolf Theorem [2] states that it is possible to compress two statistically dependent discrete random sequences and that are independently and identically distributed (i.i.d.) in a distributed way (separate encoding and joint decoding) using a rate similar to that used in a system where the sequences are encoded and decoded together, i.e., like in traditional video coding schemes....
[...]
...From the Information Theory, the Slepian–Wolf Theorem [2] states that it is possible to compress two statistically dependent discrete random sequences and that are independently and identically distributed (i....
[...]
743 citations