Content Based Retrieval of VRML Objects - An Iterative and Interactive Approach

Abstract

We examine the problem of searching a database of three-dimensional objects (given in VRML) for objects similar to a given object. We introduce an algorithm which is both iterative and interactive. Rather than base the search solely on geometric feature similarity, we propose letting the user influence future search results by marking some of the results of the current search as 'relevant' or 'irrelevant', thus indicating personal preferences. A novel approach, based on SVM, is used for the adaptation of the distant measure consistently with these markings, which brings the 'relevant' objects closer and pushes the 'irrelevant' objects farther. We show that in practice very few iterations are needed for the system to cenverge well on what the user 'had in mind'.