| Authors |
Grabner Helmut, Bischof Horst |
| Appeared in |
IEEE Conference on Computer Vision and Pattern Recognition, Volume 1 (CVPR'06), pages 260-267 |
| Date |
2006 |
| Abstract |
Boosting has become very popular in computer vision, showing impressive performance in detection and recognition tasks. In most cases off-line methods have been used for training, which implies that all training data has to be a priori given; training and application of the classifier are separate steps. However, training the classifier on-line and incrementally as new data arrives has several advantages and opens new application areas for boosting in computer vision. In this paper we propose a novel on-line AdaBoost feature selection method. Together with efficient feature extraction methods the method is real time capable. This opens many application areas in vision. We demonstrate the multifariousness of the method on such diverse tasks as
learning complex background models, visual tracking and object detection. All approaches benefit significantly by the on-line training. |
| Link |
PDF
|
IPCity Final Event @ UniAK March 24-25, 2010
Forgot your password?
