As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
The Fisher kernel is a type of kernels defined on probability distributions. It is the inner product of the Fisher score vectors, which are defined as the gradient vectors of the log likelihoods of the distributions. The main problem that arises when using the Fisher kernel for an arbitrary stochastic model is that the Fisher score vector may become very sparse, making the kernel carrying nearly no information. To cope with it, we combined the Fisher kernel with the Laplacian kernel. In the same vein, we also defined the Fisher L1 exponent kernel, using the exponential function with L1 norm. Three kernels were evaluated for ranking images. We carried out experiments using Caltech image set. The result showed that newly proposed kernels perform better than the original Fisher kernel.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.