Multimodal Similarity-Preserving Hashing

Jonathan Masci, Michael M. Bronstein, Alexander M. Bronstein, Jurgen Schmidhuber
2014 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We introduce an efficient computational framework for hashing data belonging to multiple modalities into a single representation space where they become mutually comparable. The proposed approach is based on a novel coupled siamese neural network architecture and allows unified treatment of intra-and inter-modality similarity learning. Unlike existing cross-modality similarity learning approaches, our hashing functions are not limited to binarized linear projections and can assume arbitrarily
more » ... mplex forms. We show experimentally that our method significantly outperforms state-ofthe-art hashing approaches on multimedia retrieval tasks. Index Terms-similarity-sensitive hashing, metric learning, feature descriptor J. Masci and J. Schmidhuber are with the Swiss AI Lab, Istituto Dalle Molle di Studi sull'Intelligenza Artificiale (IDSIA), Manno, Switzerland; M. M. Bronstein is with Università della Svizzera Italiana, Lugano, Switzerland and Intel Semiconductor, Switzerland; A. M. Bronstein is with the School of Electrical Engineering, Tel Aviv University, Israel and Intel Semiconductor, Israel.
doi:10.1109/tpami.2013.225 pmid:26353203 fatcat:biaqx2ztzjeyhi2n35ry7updra