The Community for Technology Leaders
Green Image
Issue No. 04 - April (2014 vol. 36)
ISSN: 0162-8828
pp: 824-830
Jonathan Masci , Swiss AI Lab. (IDSIA), Univ. of Lugano (USI), Lugano, Switzerland
Michael M. Bronstein , Inst. of Comput. Sci., Univ. of Lugano (USI), Lugano, Switzerland
Alexander M. Bronstein , Sch. of Electr. Eng., Tel Aviv Univ., Tel Aviv, Israel
Jürgen Schmidhuber , Swiss AI Lab. (IDSIA), Univ. of Lugano (USI), Lugano, Switzerland
ABSTRACT
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 complex forms. We show experimentally that our method significantly outperforms state-of-the-art hashing approaches on multimedia retrieval tasks.
INDEX TERMS
Training, Measurement, Vectors, Neural networks, Standards, Optimization, Databases
CITATION

J. Masci, M. M. Bronstein, A. M. Bronstein and J. Schmidhuber, "Multimodal Similarity-Preserving Hashing," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 36, no. 4, pp. 824-830, 2014.
doi:10.1109/TPAMI.2013.225
205 ms
(Ver 3.3 (11022016))