Zoom tracking and its applications

Abstract

Abstract. We present a new active vision technique called zoom tracking. Zoom tracking 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 visual feedback algorithm based on optical flow, and use of depth information obtained from an autofocus camera's range sensor. We explore two uses of zoom 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, such as correlation-based trackers. While zoom tracking cannot totally compensate for an object's motion, due to the effect of perspective distortion, an analysis of this distortion provides a quantitative estimate of the performance of zoom tracking. Zoom tracking can be used to reconstruct a depth map of the tracked object. We show that under normal circumstances this reconstruction is much more accurate than depth from zooming, and works over a greater range than depth from axial motion while providing, in the worst case, only slightly less accurate results. Finally, we show how zoom tracking can also be used in time-to-contact calculations.

PSO-Based SIFT False Matches Elimination for Zooming Image

“…(3) Updated the population according to equation (5) and (6). (4) The operation should stop if the termination condition is met.…”

“…According to the actual structure of lens, Asada and Baba, etc [5], presented the zoom lens model which had three parameters: zoom, focus and aperture. Fayman, etc [6], used depth estimate for vision tracking field and put forward a active vision technology of zoom tracking. However, most of these studies, lack of automatic matching algorithms, focused on 3D model reconstruction of zooming image.…”

Zooming image based false matches elimination algorithms for robot navigation

“…Model of three parameters of zoom, focus, and aperture based on the actual structure of the lens zoom is proposed by Asada and colleagues [11][12][13] The experiment results show that the model is only fit for the high accurate and low distortion lens. Active vision of zoom tracking is proposed by Fayman et al 14 for visual tracking of depth estimation in zooming images, and applications of depth estimation of zooming image have been widened. In order to realize mobile robot visual servoing for object tracking and obstacle based on zooming image, some investigations are realized in Gao et al 15 Robust feature matching based on scale-invariant feature transform (SIFT) is realized by geometry constraint of zooming image.…”

Depth from the visual motion of a planar target induced by zooming

“…The initial distance from camera to target is harder to obtain. In [2] we used a laser and other authors have proposed, for example, to use the range scanner of an autofocus camera [5], stereo correspondence, trifocal tensors [6], depth from defocused images [7] and depth from zooming.…”