Ten lectures on wavelets
Normalized Cuts and Image Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
SIMPLIcity: Semantics-Sensitive Integrated Matching for Picture LIbraries
IEEE Transactions on Pattern Analysis and Machine Intelligence
Image Browsing using Hierarchical Clustering
ISCC '99 Proceedings of the The Fourth IEEE Symposium on Computers and Communications
A Metric for Distributions with Applications to Image Databases
ICCV '98 Proceedings of the Sixth International Conference on Computer Vision
Toward bridging the annotation-retrieval gap in image search by a generative modeling approach
MULTIMEDIA '06 Proceedings of the 14th annual ACM international conference on Multimedia
Hierarchical browsing and search of large image databases
IEEE Transactions on Image Processing
Unsupervised image-set clustering using an information theoretic framework
IEEE Transactions on Image Processing
| Hi-index | 0.02 |
Content-Based Image Retrieval systems provide a variety of usages. The most common one is target search, in which a user is trying to find a specific target image. Instead, we present, in this paper, a flexible image dataset browsing system. The user can browse the whole dataset looking for any "interesting" image. To this aim, images are first abstracted through a set of signatures describing their color and texture composition. Afterwards, unsupervised clustering is performed to split the image set into several clusters of "similar" images. Every cluster is represented by its centroid as an icon. The set of icons is presented to the user, who can pick one in order to see the images belonging to the cluster. Multi-dimensional scaling is used to visualize images in the same cluster by mapping the images onto a two-dimensional space. The experiments performed with a general-purpose image dataset consisting of one thousand images, categorized into ten classes, show the usefulness of the system.