IEEE Transactions on Pattern Analysis and Machine Intelligence
ACM Computing Surveys (CSUR)
Evidence Contrary to the Statistical View of Boosting
The Journal of Machine Learning Research
Semi-supervised On-Line Boosting for Robust Tracking
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part I
Random patch based video tracking via boosting the relative spaces
ICASSP '09 Proceedings of the 2009 IEEE International Conference on Acoustics, Speech and Signal Processing
Robust tracking with discriminative ranking lists
ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part I
Robust Object Tracking with Online Multiple Instance Learning
IEEE Transactions on Pattern Analysis and Machine Intelligence
Combining discriminative and descriptive models for tracking
ACCV'09 Proceedings of the 9th Asian conference on Computer Vision - Volume Part I
Robust tracking via weakly supervised ranking SVM
CVPR '12 Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Hi-index | 0.00 |
Recently, a tracking algorithm called Multiple Instance Learning Tracker (MILTrack) has become one of the most popular methods in treating tracking problems. This technique uses an MIL based appearance model to represent training data in the form of bags. It is commonly known that during tracking, slightly inaccuracies of locations may lead to incorrectly labeled training examples, which will cause model drift problem. MILTrack is designed to alleviate this problem. In this paper, we analyze the algorithm and point out that MILTrack has a serious problem that will reduce its performance. Then a solution to this problem is proposed. Experimental results showed that our enhanced MILTrack outperformed the original one.