Representing shape with a spatial pyramid kernel

Abstract

The objective of this paper is classifying images by the ob-ject categories they contain, for example motorbikes or dol-phins. There are three areas of novelty. First, we introduce a descriptor that represents local image shape and its spa-tial layout, together with a spatial pyramid kernel. These are designed so that the shape correspondence between two images can be measured by the distance between their de-scriptors using the kernel. Second, we generalize the spatial pyramid kernel, and learn its level weighting parameters (on a validation set). This signi?cantly improves classi?cation performance. Third, we show that shape and appearance kernels may be combined (again by learning parameters on a validation set). Results are reported for classi?cation on Caltech-101 and retrieval on the TRECVID 2006 data sets. For Caltech-101 it is shown that the class speci?c optimization that we introduce exceeds the state of the art performance by more than 10%.