Maximum Correntropy Criterion for Robust Face Recognition
Citations
925 citations
Cites methods from "Maximum Correntropy Criterion for R..."
...[188] proposed utilizing the maximum correntropy criterion name d CESR embedding non-negative constraint and half-quadrati c optimization to present a robust face recognition algorith m....
[...]
744 citations
Cites background from "Maximum Correntropy Criterion for R..."
...Robust surrogate loss functions, such as the Cauchy loss (Moore 1977) function and correntropy (Welsch loss function) (Liu et al 2007; He et al 2011) have been empirically proven to be robust to noise....
[...]
420 citations
355 citations
327 citations
Cites methods from "Maximum Correntropy Criterion for R..."
...Provided that the is orthonormal, i.e., , we can obtain (9) Substituting (projection theorem [19]) into (7) and according to (9), we get the following optimization problem: (10) where is an orthonormal matrix and ....
[...]
References
45,034 citations
Additional excerpts
...Ç...
[...]
40,147 citations
"Maximum Correntropy Criterion for R..." refers background in this paper
...where k ð:Þ is a kernel function that satisfies Mercer theory [40] and E1⁄2: is the expectation operator....
[...]
[...]
33,341 citations
14,562 citations
"Maximum Correntropy Criterion for R..." refers methods in this paper
...Ç...
[...]
...Compared with the state-of-the-art l1norm-based sparse representation classifier (SRC), which assumes that noise also has a sparse representation, our sparse algorithm is developed based on the maximum correntropy criterion, which is much more insensitive to outliers....
[...]
11,674 citations
"Maximum Correntropy Criterion for R..." refers methods or result in this paper
...Along this line, Wright et al. [24] recently proposed a sparse representation classifier (SRC) for robust face recognition against occlusions and corruptions, and impressive results were reported as compared to many wellknown face recognition methods [1], [ 2 ], [25]....
[...]
...Well-known face recognition algorithms such as Eigenface [1], Fisherface [ 2 ], and Independent face [3] do not explicitly and effectively consider extracting robust facial features, although they are very useful for extracting descriptive or discriminative features....
[...]