Foveal images have variable spatial resolution, enabling a significant reduction in image size and data volume. Recently, a new class of active foveated vision system was proposed [1]. This paper examines the performance of this system for tracking a single target. It demonstrates that in the case of static targets, the fovea is able to be positioned within 1 pixel of the true target location within two frames. In the case of dynamic targets, it is demonstrated that there is a significant improvement over using a uniform low resolution image with the same number of pixels, where using a foveated image gives only a slight increase in tracking error compared with tracking the target using a high resolution image.