Kernel Coding: General Formulation and Special Cases

TL;DR

This paper introduces a unified kernel coding framework that performs image coding in high-dimensional Hilbert spaces, enhancing class separability and improving visual recognition performance.

Contribution

It presents a general kernel coding formulation encompassing popular methods and explores joint learning of kernels, dictionaries, and classifiers.

Findings

Kernel coding in Hilbert space improves class separability.

Joint learning of kernel, dictionary, and classifier yields better recognition results.

Experimental results demonstrate the effectiveness of the proposed approach.

Abstract

Representing images by compact codes has proven beneficial for many visual recognition tasks. Most existing techniques, however, perform this coding step directly in image feature space, where the distributions of the different classes are typically entangled. In contrast, here, we study the problem of performing coding in a high-dimensional Hilbert space, where the classes are expected to be more easily separable. To this end, we introduce a general coding formulation that englobes the most popular techniques, such as bag of words, sparse coding and locality-based coding, and show how this formulation and its special cases can be kernelized. Importantly, we address several aspects of learning in our general formulation, such as kernel learning, dictionary learning and supervised kernel coding. Our experimental evaluation on several visual recognition tasks demonstrates the benefits of performing coding in Hilbert space, and in particular of jointly learning the kernel, the dictionary and the classifier.