Using Visual-Textual Mutual Information and Entropy for Inter-modal Document Indexing.

Jean Martinet, Shin'ichi Satoh

Browse the full ECIR paper archive.