Abstract
Ob ject detection is one of the key problems in computer vi- sion. In the last decade, discriminative learning approaches have proven effective in detecting rigid ob jects, achieving very low false positives rates. The field has also seen a resurgence of part-based recognition methods, with impressive results on highly articulated, diverse ob ject categories. In this paper we propose a discriminative learning approach for detection that is inspired by part-based recognition approaches. Our method, Multiple Component Learning (mcl), automatically learns indi- vidual component classifiers and combines these into an overall classifier. Unlike previous methods, which rely on either fairly restricted part mod- els or labeled part data, mcl learns powerful component classifiers in a weakly supervised manner, where ob ject labels are provided but part la- bels are not. The basis of mcl lies in learning a set classifier; we achieve this by combining boosting with weakly supervised learning, specifically the Multiple Instance Learning framework (mil). mcl is general, and we demonstrate results on a range of data from computer audition and computer vision. In particular, mcl outperforms all existing methods on the challenging INRIA pedestrian detection dataset, and unlike methods that are not part-based, mcl is quite robust to occlusions.