Universal Multimedia Access and
Semantic Summarization for Presentations

Multimedia content is being both disseminated and consumed on a larger scale than ever before. In particular, the growth of collaborative applications (e.g. Distance Learning (DL)) means that multimedia-rich audio-visual presentations are increasingly available on the Internet. At the same time, mobile devices capable of processing multimedia applications have gained popularity. Consequently, designers of multimedia systems face two major challenges driven by these factors: First, users require access to personalized content seamlessly over differing networks and on virtually any terminal device. Second, users require a means of navigating and digesting the vast content efficiently.

This work proposes a presentation summarization framework for audio-visual presentations in a DL environment which addresses the aforementioned challenges. The framework is designed based on emerging international standards to provide a Universal Multimedia Access (UMA) solution that is able to deliver content to heterogeneous users. A novel summarization scheme is proposed which combines information from multiple sources such as audio data and presentation slides to improve accessibility of presentation archives for users.

Note that preliminary work for this project was done under the project Time Stretching an Audio Signal.

PPTX Parser Technical Report »

Team Members: Mohammed Ajmal, Azadeh Kushki

Related Publications: