Tracking with Online Multiple Instance Learning

In this paper we address the problem of tracking an object in a video given its location in the first frame and no other information. Recently, a class of tracking techniques called "tracking by detection" has been shown to give promising results at real-time speeds. These methods train a discriminative classifier in an online manner to separate the object from the background. This classifier bootstraps itself by using the current tracker state to extract positive and negative examples from the current frame. Slight inaccuracies in the tracker can therefore lead to incorrectly labeled training examples, which degrade the classifier and can cause further drift. In this paper we show that using Multiple Instance Learning (MIL) instead of traditional supervised learning avoids these problems, and can therefore lead to a more robust tracker with fewer parameter tweaks. We propose a novel online MIL algorithm for object tracking that achieves superior results with real-time performance. We present thorough experimental results (both qualitative and quantitative) on a number of challenging video clips.

Figure 1 - Updating a discriminative appearance model: (A) Using a single positive image patch to update a traditional discriminative classifier. The positive image patch chosen does not capture the object perfectly. (B) Using several positive image patches to update a traditional discriminative classifier. This can confuse the classifier causing poor performance. (C) Using one positive bag consisting of several image patches to update a MIL classifier.

related publications

@inproceedings {babenko11,
title = {Robust Object Tracking with Online Multiple Instance Learning},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
year = {2011},
author = {Boris Babenko and Ming-Hsuan Yang Serge Belongie}
}

@inproceedings{babenko_cvpr09,
title=,
author={B. Babenko and Ming-Hsuan Yang and S. Belongie},
booktitle={CVPR},
year={2009},
}

code

data

For each clip we provide a zip file that contains the following: (1) a directory with the original image sequence; image are named "img0000.png", "img00001.png", etc. (2) a [name]_frames.txt file that contains the frame number of the first and last frame of the sequence, (3) a [name]_gt.txt file that contains ground truth object locations; each line corresponds to a frame, and contains the "x,y,width,height"; note that this information is only available for 1 in every 5 frames, the rest is filled with 0's, (4) MILTrack restults for 5 trials in the same format as above.