We present a new active vision technique called zoom tracking. Zoom trackin
g is the continuous adjustment of a camera's focal length in order to keep
a constant-sized image of an object moving along the camera's optical axis.
Two methods for performing zoom tracking are presented: a closed-loop visu
al feedback algorithm based on optical flow, and use of depth information o
btained from an autofocus camera's range sensor. We explore two uses of zoo
m tracking: recovery of depth information and improving the performance of
scale-variant algorithms. We show that the image stability provided by zoom
tracking improves the performance of algorithms that are scale variant, su
ch as correlation-based trackers. While zoom tracking cannot totally compen
sate for an object's motion, due to the effect of perspective distortion, a
n analysis of this distortion provides a quantitative estimate of the perfo
rmance of zoom tracking. Zoom tracking can be used to reconstruct a depth m
ap of the tracked object. We show that under normal circumstances this reco
nstruction is much more accurate than depth from zooming, and works over a
greater range than depth from axial motion while providing, in the worst ca
se, only slightly less accurate results. Finally, we show how zoom tracking
can also be used in time-to-contact calculations.