Robust discrete code modeling for supervised hashing

Yadan Luo, Yang Yang, Fumin Shen, Zi Huang, Pan Zhou, Heng Tao Shen
2018 Pattern Recognition  
Highlights • We propose a novel supervised hashing scheme to generate high-quality hash codes and hash functions for facilitating large-scale multimedia applications. • We devise an effective binary code modeling approach based on l2,p-norm, which can adaptively induce sample-wise sparsity, to perform automatic code selection as well as noisy samples identification. • We preserve the discrete constraint in the proposed model to directly producediscrete codes with minimal quantization error. An
more » ... fficient algorithms is designed to solve the discrete optimization problem, where a weighted discrete cyclic coordinate decent (WDCC) algorithm is proposed to derive robust binary codes. • Extensive experiments conducted on various real-world datasets demonstrate the promising results of the RDCM approach in retrieval and classification tasks. Abstract Recent years have witnessed the promising efficacy and efficiency of hashing (also known as binary code learning) for retrieving nearest neighbor in large-scale data collections. Particularly, with supervision knowledge (e.g., semantic labels), we may further gain considerable performance boost. Nevertheless, most existing supervised hashing schemes suffer from the following limitations: 1) severe quantization error caused by continuous relaxation of binary codes; 2) disturbance of unreliable codes in subsequent hash function learning; and 3) erroneous guidance derived from imprecise and incomplete semantic labels. In this work, we propose a novel supervised hashing approach, termed as Robust Discrete Code Modeling (RDCM), which directly learns high-quality discrete binary codes and hash functions by effectively suppressing the influence of unreliable binary codes and potentially noisily-labeled samples. RDCM employs 2,p norm, which is capable of inducing sample-wise sparsity, to jointly perform code selection and noisy sample identification. Moreover, we preserve the discrete constraint in RDCM to eliminate the quantization error. An efficient algorithm is developed to solve the discrete optimization problem. Extensive experiments conducted on various real-life datasets show the superiority of the proposed RDCM approach as compared to several state-of-the-art hashing methods.
doi:10.1016/j.patcog.2017.02.034 fatcat:5n2xtxvg7jh7pofrbcahv33ofm