Fuzzy object patterns for visual indexing and segmentation

In this paper, we proposed Fuzzy Object Pattern (FOP) for the represention of image content. FOP is derived from view-based object recognition against a predefined vocabulary of visual object classes. Tessellation of FOPs over an image is further aggregated spatially to summarize the image content. This description scheme is deployed in image indexing and retrieval on home photographs with very promising results. Furthermore, FOP spans a new fuzzy pattern space in which incremental clustering is carried out to aggregate adjacent FOPs into larger regions. As a consequence, dominant regions can be segmented from an image.