Brownian Descriptor: A Rich Meta-Feature for Appearance Matching

Abstract

This paper introduces an image region descriptor and applies it to the problem of appearance matching. The pro-posed descriptor can be seen as a natural extension of co-variance. Driven by recent studies in mathematical statis-tics related to Brownian motion, we design the Brownian descriptor. In contrast to the classical covariance descrip-tor, which measures the degree of linear relationship be-tween features, our novel descriptor measures the degree of all kinds of possible relationships between features. We argue that the proposed covariance is a richer descrip-tor than the classical covariance, especially when fusing non-linearly dependent features. We evaluate our approach on tracking related applications, demonstrating that the Brownian descriptor outperforms the classical covariance in terms of matching accuracy and efficiency. 1.