Humoristische Auseinandersetzung mit den Jugendschutzmaßnahmen von Apple“
— Markus Beckendahl,


johannes p osterhoff - semex

semex is a system for semantic video retrieval.
Like the visionary memex it strives for explorative and meaningful information retrieval.

The Semantic Multimedia Explorer research project was initiated by Hasso-Plattner-Institute in Potsdam to develop technologies that reply to the growing amount of multimedia content on the World Wide Web. The Semantic Media Explorer, or semex as we call it, combines state-of-the-art media analysis techniques, such as structural segmentation, intelligent character recognition, the ability to detect genres and faces in videos and semantic technologies. In this project I am responsible for the interface design of semex.

Capabilites of the semex

The amount of multimedia content on the World Wide Web is growing rapidly. To manage and organize this content new technologies are required. One example of such technologies is the semantic multimedia search. Semantic multimedia search offers an efficient way to access this growing amount of data. Automated processes for semantic analysis of audiovisual content provide the solution to the current problem.

The Semantic Media Explorer combines state-of-theart media analysis procedures, such as automatic scene segmentation, intelligent character recognition, and the ability to detect genres and faces in videos. As a result, the Semantic Media Explorer provides optimal access to video content. This new kind of multimedia search offers users enhanced support with semantically enhanced metadata and intuitive user interfaces. The information extracted via the media analysis is correlated using semantic information so that users can identify new links and rummage around in the multimedia content.

Read more about semex in the official PDF from Hasso Plattner Institute.