Interactive Hyperbolic Image Browsing – Towards an Integrated Multimedia Navigator

Search in and presentation of multimedia collections is an important and complex task. Often it involves various search strategies and the visualization of found object collections. Technically, search strategies can be formulated using distance measures between object pairs. In this paper we propose the use of a projection based method to dynamically arrange a set of objects (here images) given a notion of image similarity (an interactive user choice). This approach uses the previously introduced Hyperbolic Multidimensional Scaling method (HMDS) in order to find spatial layouts of the objects in the hyperbolic plane IH based on pairwise dissimilarity data. The circular Poincaré model of the IH allows effective human-computer interaction: by moving the “focus” via mouse the user can navigate in the images without loosing the “context” around the center, which appears with gradually lower resolution. The exponential growth of area around each point in the IH makes this non-Euclidean projection space extraordinary for the layout also for images. We use the HMDS technique to interactively mix various concepts of “image similarity” (based on visual and annotation information). Depending on the preferences and the actual task the user can modulate the distance metric while observing the resulting rearrangement of the images in IH. The concept can be generalized to any kind of multimedia content, given suitable similarity functions on the content, e.g. distances in ontologies, sound features, or multimedia descriptions etc.

[1]  Michael Gervautz,et al.  A simple method for color quantization: octree quantization , 1990 .

[2]  Ramana Rao,et al.  Laying out and visualizing large trees using a hyperbolic space , 1994, UIST '94.

[3]  Ramana Rao,et al.  A focus+context technique based on hyperbolic geometry for visualizing large hierarchies , 1995, CHI '95.

[4]  B. S. Manjunath,et al.  Texture Features for Browsing and Retrieval of Image Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Frank Morgan,et al.  Riemannian geometry : a beginner's guide , 1998 .

[6]  Shih-Fu Chang,et al.  Image Retrieval: Current Techniques, Promising Directions, and Open Issues , 1999, J. Vis. Commun. Image Represent..

[7]  Carlo Tomasi,et al.  Perceptual metrics for image database navigation , 1999 .

[8]  H. Ritter.  Self-Organizing Maps on non-euclidean Spaces , 1999 .

[9]  Mary Czerwinski,et al.  An initial examination of ease of use for 2D and 3D information visualizations of web content , 2000, Int. J. Hum. Comput. Stud..

[10]  Ingemar J. Cox,et al.  The Bayesian image retrieval system, PicHunter: theory, implementation, and psychophysical experiments , 2000, IEEE Trans. Image Process..

[11]  Helge J. Ritter,et al.  On interactive visualization of high-dimensional data using the hyperbolic plane , 2002, KDD.

[12]  Helge J. Ritter,et al.  Interactive visualization and navigation in large data collections using the hyperbolic space , 2003, Third IEEE International Conference on Data Mining.

[13]  G. Sagerer,et al.  Content-based image retrieval by multimodal interaction , 2003, IECON'03. 29th Annual Conference of the IEEE Industrial Electronics Society (IEEE Cat. No.03CH37468).

[14]  Jörg A. Walter.  H-MDS: a new approach for interactive visualization with multidimensional scaling in the hyperbolic space , 2004, Inf. Syst..