Learning Image Descriptors with the Boosting-Trick
T. P. Trzcinski, Mario Christoudias, Vincent Lepetit, Pascal Fua · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2012
In this paper we apply boosting to learn complex non-linear local visual feature representations, drawing inspiration from its successful application to visual ob-ject detection. The main goal of local feature descriptors is to distinctively repre-sent a salient image region while remaining invariant to viewpoint and illumina-tion changes. This representation can be improved using machine learning, how-ever, past approaches have been mostly limited to learning linear feature mappings in either the original input or a kernelized input feature space. While kernelized methods have proven somewhat effective for learning non-linear local feature de-scriptors, they rely heavily on the choice of an appropriate kernel function whose selection is often difficult and non-intuitive. We propose to use the boosting-trick to obtain a non-linear mapping of the input to a high-dimensional feature space. The non-linear feature mapping obtained with the boosting-trick is highly intu-itive. We employ gradient-based weak learners resulting in a learned descriptor that closely resembles the well-known SIFT. As demonstrated in our experiments, the resulting descriptor can be learned directly from intensity patches achieving state-of-the-art performance. 1