Multimedia Knowledge Exploitation for E-Learning: Some Enabling Techniques

Knowledge is embedded in multimedia either explicitly or implicitly. In a practical application like e-learning, more than one type of media will take effect. In this paper, different ways of feature extraction from multimedia are explored. The media being analyzed and retrieved are not only images, but video, audio and even 3D terrains as well. New algorithms and experimental results are presented. As a result of the integration of multi-modal media, we lay down a foundation for exploiting media knowledge effectively, which can greatly enhance the performance of the high-level semantic retrieval desired by advanced applications such as e-learning.