Object Classification

Object detection and classification has intensely been researched in computer vision [#!CVAMA!#,#!Papageorgio_1998!#,#!Rowley_1998!#,#!Viola_2001!#]. Common approaches use for example neural networks or support vector machines (SVM) to detect and classify objects. Rowley et al. detect faces using a small set of simple features and neural networks [#!Rowley_1998!#] and Papageorgiou et al. recognize pedestrians with simple vertical, horizontal and diagonal features and SVMs [#!Papageorgio_1998!#]. Recently Viola and Jones have proposed a boosted cascade of simple classifiers for fast face detection [#!Viola_2001!#]. Inspired by these ideas, we detect objects, e.g., office chairs [#!MPI!#], in 3D range and reflectance data using a cascade of classifiers composed of several simple classifiers, which in turn contain an edge, line or center surround feature.

Figure: Left: Edge, line, diagonal and center surround features are used for classification. Right: Computation of feature values in the shaded region is based on the four upper rectangles.