Copyright © 2014 by Intelligent Systems Laboratory, Computer Science Department, Technion - Israel Institute of Technology, Haifa 3200003, Israel. All rights reserved

Bittracker - A Bitmap Tracker for Visual Tracking Under Very General Conditions

Bittracker - A Bitmap Tracker for Visual Tracking Under Very General ConditionsBy: Ido Leichter, Michael Lindenbaum and Ehud Rivlin

This paper addresses the problem of visual tracking under very general conditions: a possibly non-rigid target whose appearance may drastically change over time; general camera motion; a 3D scene; and no a priori information except initialization. This is in contrast to the vast majority of trackers which rely on some limited model in which, for example, the target's appearance is known a priori or restricted, the scene is planar, or a pan tilt zoom camera is used. Their goal is to achieve speed and robustness, but their limited context may cause them to fail in the more general case. 



The proposed tracker works by approximating, in each frame, a PDF (probability distribution function) of the target's bitmap and then estimating the maximum a posteriori bitmap. The PDF is marginalized over all possible motions per pixel, thus avoiding the stage in which optical flow is determined. This is an advantage over other general-context trackers that do not use the motion cue at all or rely on the error-prone calculation of optical flow. Using a Gibbs distribution with a first-order neighborhood system yields a bitmap PDF whose maximization may be transformed into that of a quadratic pseudo-Boolean function, the maximum of which is approximated via a reduction to a maximum-flow problem. Many experiments were conducted to demonstrate that the tracker is able to track under the aforementioned general context.

Bittracker - A Bitmap Tracker for Visual Tracking Under Very General Conditions


To appear in IEEE Transaction on pattern Analysis and Machine Intelligence.

Paper download: PDF


Experimental results

Cellotape Sequence: The hole in the reel, which is not revealed at the beginning of the video, is revealed and marked correctly as the video progresses.

Boat Sequence: The background motion is caused not only by the camera motion, but also by the motion of the water.

Herd Sequence: The tracking overcomes a severe partial occlusion.

Lighter Sequence: The areas of the lighter that were previously occluded by other objects or by the lighter itself are correctly classified upon exposure.

Man-in-Mall Sequence: Although parts of the target are occasionally misclassified, these are corrected with time due to the spatial motion continuity and the spatial color coherence assumptions. Note the zoom-in and zoom-out near the end of the sequence, and the partial occlusion at the end.

Woman-and-Child Sequence: The tracking overcomes lighting changes and long-term partial occlusions. Since the woman and the girl she takes by the hand are adjacent and walking at similar velocity over an extended time period, the girl is joined to the woman in the tracking process.

Ball Sequence: The areas of the ball that appear during its roll behind the toy are correctly classified

Random-dot Sequence: Tracking a random-dot object of gradually time-varying shape and colors moving in front of a random-dot background of gradually time-varying colors that is in motion as well. It is evident that the tracking in this sequence is very accurate. Note that new object pixels and revealed background pixels are correctly classified, due to the spatial motion continuity assumption.

A random-dot object moving in front of a random-dot background that is moving as well.

The random-dot object with the background cut out.

The estimated bitmap is shown in green, overlaid on top of a grayscale version video containing only the target.