Munehiro Nakazato and Thomas S. Huang,
3D MARS: Immersive Virtual Reality for Content-Based Image
Retrieval,
In Proceedings of the IEEE International Conference on
Multimedia and Expo (ICME2001),
Tokyo, Japan, October 22-25 2001.
Munehiro Nakazato and Thomas S. Huang,
3D MARS: Immersive Virtual Reality for Content-Based Image
Retrieval,
In Proceedings of the IEEE International Conference on
Multimedia and Expo (ICME2001),
Tokyo, Japan, October 22-25 2001.3D MARS is a new visualization system for Content-Based Image Retrieval (CBIR) based on the MARS project of University of Illinois. In 3D MARS, the user browses and queries images in an immersive 3D Virtual Reality space of CAVE. The results of the query are displayed in 3D, so that the user can see the result with respect to three different criteria such as color, texture and structure. The user can examine the result from different view angles by flying-through the space with the joystick. By giving feedback to the previous query results, the use can incrementally improve the query (relevance feedback.) By the use of both Feature Ranking and Total Ranking, the system makes the retrieval process more flexible. In addition, the Sphere mode visualization is provided as a powerful analyzing tool for CBIR researchers. Although the system is developed on an immersive projection-based virtual reality environment, our approach can also be used on different types of virtual reality system such as Head-Mounted Display or monitor-based VR.