@inproceedings{5259,
  title = {A Multiple Kernel Learning Approach to Joint Multi-Class Object Detection},
  journal = {Pattern Recognition: Proceedings of the 30th DAGM Symposium},
  booktitle = {DAGM 2008},
  abstract = {Most current methods for multi-class object classification
  and localization work as independent 1-vs-rest classifiers. They decide
  whether and where an object is visible in an image purely on a per-class
  basis. Joint learning of more than one object class would generally be
  preferable, since this would allow the use of contextual information such
  as co-occurrence between classes. However, this approach is usually not
  employed because of its computational cost.
  In this paper we propose a method to combine the efficiency of single
  class localization with a subsequent decision process that works jointly
  for all given object classes. By following a multiple kernel learning (MKL)
  approach, we automatically obtain a sparse dependency graph of relevant
  object classes on which to base the decision. Experiments on the
  PASCAL VOC 2006 and 2007 datasets show that the subsequent joint
  decision step clearly improves the accuracy compared to single class
  detection.},
  pages = {31-40},
  editors = {Rigoll, G. },
  publisher = {Springer},
  organization = {Max-Planck-Gesellschaft},
  institution = {German Association for Pattern Recognition},
  school = {Biologische Kybernetik},
  address = {Berlin, Germany},
  month = jun,
  year = {2008},
  note = {Main Award DAGM 2008},
  slug = {5259},
  author = {Lampert, C. and Blaschko, MB.},
  month_numeric = {6}
}