Towards Semantic and Effective Visual Codebooks

This book focuses on the study of visual vocabularies for category-level object recognition. Our aim is not just to obtain more discriminative and more compact visual codebooks, but to bridge the gap between visual features and semantic concepts. A novel approach for obtaining class representative visual words is presented. It is based on a maximisation procedure, i.e. the Cluster Precision Maximisation, of a novel cluster precision criterion, and on an adaptive threshold refinement scheme for agglomerative clustering algorithms based on correlation clustering techniques. A novel clustering aggregation based approach for building effective visual vocabularies is described too. It consist of a novel framework for incorporating meaningful spatial coherency among the local features into the visual codebook construction. We also propose an efficient high-dimensional data clustering algorithm, the Fast Reciprocal Nearest Neighbours. Finally, we release a new database of images called Image Collection of Annotated Real-world Objects (ICARO), which is especially designed for evaluating category-level object recognition systems.

Roberto López-Sastre received the M.Sc. and the Ph.D. degrees in Electrical Engineering from the University of Alcalá, Spain in 2005 and 2010. He is currently Assistant Professor at the Department of Signal Theory and Communications of the University of Alcalá. His research interests include category-level object recognition and pose estimation.