Semi-Supervised Nonlinear Hashing Using Bootstrap Sequential Projection Learning

Citations Over TimeTop 10% of 2012 papers

Abstract

In this paper, we study the effective semi-supervised hashing method under the framework of regularized learning-based hashing. A nonlinear hash function is introduced to capture the underlying relationship among data points. Thus, the dimensionality of the matrix for computation is not only independent from the dimensionality of the original data space but also much smaller than the one using linear hash function. To effectively deal with the error accumulated during converting the real-value embeddings into the binary code after relaxation, we propose a semi-supervised nonlinear hashing algorithm using bootstrap sequential projection learning which effectively corrects the errors by taking into account of all the previous learned bits holistically without incurring the extra computational overhead. Experimental results on the six benchmark data sets demonstrate that the presented method outperforms the state-of-the-art hashing algorithms at a large margin.

Related Papers

• → Compact hashing for mixed image-keyword query over multi-label images(2012)21 cited
• → Grassmann Hashing for approximate nearest neighbor search in high dimensional space(2011)15 cited
• → Learning to Hash on Structured Data(2015)8 cited
• → Cluster-based data oriented hashing(2015)1 cited
• → Eclipse Hashing: Alexandrov Compactification and Hashing with Hyperspheres for Fast Similarity Search(2014)