Multimodal Discriminative Binary Embedding for Large-Scale Cross-Modal Retrieval
Authors:
- 18 July 2016
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 25 (10) , 4540-4554
Abstract
Multimodal hashing, which conducts effective and efficient nearest neighbor search across heterogeneous data on large-scale multimedia databases, has been attracting increasing interest, given the explosive growth of multimedia content on the Internet. Recent multimodal hashing research mainly aims at learning the compact binary codes to preserve semantic information given by labels. The overwhelming majority of these methods are similarity preserving approaches which approximate pairwise similarity matrix with Hamming distances between the to-be-learnt binary hash codes. However, these methods ignore the discriminative property in hash learning process, which results in hash codes from different classes undistinguished, and therefore reduces the accuracy and robustness for the nearest neighbor search. To this end, we present a novel multimodal hashing method, named multimodal discriminative binary embedding (MDBE), which focuses on learning discriminative hash codes. First, the proposed method formulates the hash function learning in terms of classification, where the binary codes generated by the learned hash functions are expected to be discriminative. And then, it exploits the label information to discover the shared structures inside heterogeneous data. Finally, the learned structures are preserved for hash codes to produce similar binary codes in the same class. Hence, the proposed MDBE can preserve both discriminability and similarity for hash codes, and will enhance retrieval accuracy. Thorough experiments on benchmark data sets demonstrate that the proposed method achieves excellent accuracy and competitive computational efficiency compared with the state-of-the-art methods for large-scale cross-modal retrieval task.Keywords
Funding Information
- National Natural Science Foundation of China (61432014, 61472304, 61501349)
- Key Industrial Innovation Chain in Industrial Domain (2016KTZDGY-02)
- Program for Changjiang Scholars and Innovative Research Team in University of China (IRT13088)
- Shaanxi Innovative Research Team for Key Science and Technology (2012KCT-02)
This publication has 32 references indexed in Scilit:
- Bit-Scalable Deep Hashing With Regularized Similarity Learning for Image Retrieval and Person Re-IdentificationIEEE Transactions on Image Processing, 2015
- Neighborhood Discriminant Hashing for Large-Scale Image RetrievalIEEE Transactions on Image Processing, 2015
- Robust Structured Subspace Learning for Data RepresentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- Multiview Alignment Hashing for Efficient Image SearchIEEE Transactions on Image Processing, 2015
- On the Role of Correlation and Abstraction in Cross-Modal Multimedia RetrievalIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Effective Multiple Feature Hashing for Large-Scale Near-Duplicate Video RetrievalIEEE Transactions on Multimedia, 2013
- Clustering-Guided Sparse Structural Learning for Unsupervised Feature SelectionIEEE Transactions on Knowledge and Data Engineering, 2013
- Image Transformation Based on Learning Dictionaries across Image SpacesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2012
- Composite hashing with multiple information sources2011
- The MIR flickr retrieval evaluation2008