Content Discovery and Retrieval Platform

In this task, we expect to provide an important contribution in the field of information retrieval, as also strengthen the collaboration between partners (BMD and Leibniz-IPHT), which have a thorough knowledge of cross-modality data analysis and data fusion. The main output will be an extensible architecture for multimodal information retrieval, supporting new algorithms without major changes to the software, thus providing a multimodal layer of abstraction over the large domain of existing retrieval algorithms, such as feature extractors and model representations. Automated content discovery mechanism will be considered beneficial: the automatic detection of complex patterns in visual content introduces new concepts in medical image retrieval. In light of these ideas, it will be studied the subject of automated object recognition in spectroscopic data and medical images, followed by the design of an architecture for information enrichment of integrated archives through automatic extraction and indexing of visual information.