Abstract

Direct visual tracking can be impaired by changes in illumi- nation if the right choice of similarity function and photometric model is not made. Tracking using the sum of squared differences, for instance, often needs to be coupled with a photometric model to mitigate illumi- nation changes. More sophisticated similarities, e.g. mutual information and cross cumulative residual entropy, however, can cope with complex illumination variations at the cost of a reduction of the convergence ra- dius, and an increase of the computational effort. In this context, the normalized cross correlation (NCC) represents an interesting alterna- tive. The NCC is intrinsically invariant to affine illumination changes, and also presents low computational cost. This article proposes a new direct visual tracking method based on the NCC. Two techniques have been developed to improve the robustness to complex illumination vari- ations and partial occlusions. These techniques are based on subregion clusterization, and weighting by a residue invariant to affine illumination changes. The last contribution is an efficient Newton-style optimization procedure that does not require the explicit computation of the Hessian. The proposed method is compared against the state of the art using a benchmark database with ground-truth, as well as real-world sequences.