Traditional multimedia indexing methods are based on the principle of hierarchical clustering of the data space, in which metric properties are used to build a tree that then can be used to prune branches while processing the queries. However, the performance of these methods will deteriorate rapidly when the dimensionality of the data space is increased.

Based on the generic multimedia indexing (GEMINI) approach and lower bounding methods a hierarchical linear subspace indexing method will be described, which does not suffer from the dimensionality problem. The hierarchical subspace approach offers a fast searching method for large content-based multimedia databases.

The approach will be demonstrated on image indexing, in which the subspaces correspond to different resolutions of the images. During content-based image retrieval the search starts in the subspace with the lowest resolution of the images. In this subspace the set off all possible similar images is determined. In the next subspace additional information corresponding to a higher resolution is used to reduce this set. This procedure is repeated until the similar images can be determined eliminating the false candidates.

The developed methods of analysis can be generalized for all means of content-based access methods that are based on information loss techniques, like for example hierarchical clustering which relies on stepwise digitalization of the space rather then the reduction of its dimension.