Max-margin additive classifiers for detection
- 1 September 2009
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 9 (15505499) , 40-47
- https://doi.org/10.1109/iccv.2009.5459203
Abstract
We present methods for training high quality object detectors very quickly. The core contribution is a pair of fast training algorithms for piece-wise linear classifiers, which can approximate arbitrary additive models. The classifiers are trained in a max-margin framework and significantly outperform linear classifiers on a variety of vision datasets. We report experimental results quantifying training time and accuracy on image classification tasks and pedestrian detection, including detection results better than the best previous on the INRIA dataset with faster training.Keywords
This publication has 29 references indexed in Scilit:
- A dual coordinate descent method for large-scale linear SVMPublished by Association for Computing Machinery (ACM) ,2008
- Conditionally Positive Definite Kernels for SVM Based Image RecognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Names and faces in the newsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Online Learning with KernelsIEEE Transactions on Signal Processing, 2004
- Robust Real-Time Face DetectionInternational Journal of Computer Vision, 2004
- Fast pose estimation with parameter-sensitive hashingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A Tale of Two Classifiers: SNoW vs. SVM in Visual RecognitionPublished by Springer Nature ,2002
- Learning Additive Models Online with Fast Evaluating KernelsPublished by Springer Nature ,2001
- Support-vector networksMachine Learning, 1995
- Generalized Additive ModelsStatistical Science, 1986