This activity is related to a reference end-to-end multimedia search and retrieval architecture which will allow efficient implementation in different environments. The goal of this exploration activity is twofold, addressing both image and video multimedia search engines. For this reason, the exploration activity is subdivided in two different tasks:
In the image multimedia search engine domain, the goal of the research is to develop a fast and robust framework for automatic image replica and digital right infringements detection. The application of the framework to automatic copyright violation in the World Wide Web will be also addressed;
In the video multimedia search engine domain, the integrated activity aims at investigating the combined usage of low and high level features automatically extracted from visual contents with other kinds of metadata available, to improve the performance of a search and retrieval system. Examples of this kind of metadata are EXIF information, GPS coordinates, textual descriptions and ratings provided by users.
Due to the nature of this explorative work, the partners involved in the research activity are contributing to the on-going MPEG query format (MPQF) Standardization initiative and also to the JPSearch standard.