This paper aims at introducing a kernel for histograms of visual words, namely the PQ kernel. ... Kernels for Visual Words Histograms. In: Petrosino, A. (eds) ...

This paper aims at introducing a kernel for histograms of visual words, namely the PQ kernel. This kernel is inspired from a class of similarity measures for ...

In this work, a kernel for histograms of visual words, namely the PQ kernel, ... All common kernels used in computer vision treat histograms of visual words.

Computer vision researchers have developed several learning methods based on the bag-of-words model for image related tasks, such as image retrieval or ...

The PQ kernel has the best results on both data sets, even when a spatial pyramid representation is used. In conclusion, the PQ kernel can be used to obtain a ...

We performed k-means clustering of the random LARK descriptors from the training images to build a visual vocabulary. It is named LARK vocabulary in this paper.

In this work, we highlight the kernel view of orientation histograms and provide a unified way to low-level image feature design and learning. Our low-level ...

LARK is used to describe the image local information and build the visual vocabulary. Each pixel of an image is assigned to the visual words and gets the ...

We propose a histogram kernel k-means algorithm which is easy to implement and runs almost as fast as the standard k-means. The HIK codebooks have consistently ...

These kernels are typically much better adapted to data representations in the form of finite probability distributions or normalized histograms, that common ...