Multi-form hierarchical representation of image categories for browsing and retrieval

In this paper, we present a multi-form image representation and adaptive weighting approach for image browsing and retrieval systems. To support the development of the multi-form representation with understanding on users´ image searching behavior, we conducted an interview study with users of various image browsing and retrieval systems. Based on the insights gained from the user study we propose an adaptive weighting method for multi-form structures. The proposed adaptive weighting approach aims at improving the efficiency and accuracy of the system. Users´ behaviors are modeled based on statistics of the past actions and weights of the forms are updated adaptively on iterations of browsing and retrieval. The proposed method has been evaluated and results are illustrated in the paper. It is shown that, satisfactory improvements can be achieved with adaptive weighting approach in the multi-form scheme.