On-line inverse multiple instance boosting for classifier grids

Article ID	Journal	Published Year	Pages	File Type
10362089	Pattern Recognition Letters	2012	8 Pages	PDF

Abstract

âº Classifier grids showed excellent detection results for stationary cameras. âº On-line adaptive classifiers reduce the complexity of the classification task. âº Fixed update strategies allow long-term stability. âº Short-term stability through proposed inverse multiple instance learning.

Keywords

AdaBoost On-line learning Object detection Multiple instance learning

Related Topics

Physical Sciences and Engineering Computer Science Computer Vision and Pattern Recognition

Preview

On-line inverse multiple instance boosting for classifier grids

Authors

Sabine Sternig, Peter M. Roth, Horst Bischof,