A new method is presented to learn object categories from unlabeled and unsegmented images for generic object recognition. We assume that each object can be characterized by a set of typical regions, and use a new segmentation method - "similarity-measure segmentation" - to split the images into regions of interest. This approach may also deliver segments, which are split into several disconnected parts, which turn out to be a powerful description of local similarities. Several textural features are calculated for each region, which are used to learn object categories with boosting. We demonstrate the flexibility and power of our method by excellent results on various datasets. In comparison, our recognition results are significantly higher than the results published in related work.