Weakly-Supervised Enhanced Semantic-Aware Hashing for Cross-Modal Retrieval

Abstract

Owing to its query and storage efficiency, hash learning has sparked much interest for Cross-Modal Retrieval (CMR). Previous literatures have proved the superiority of supervised Cross-Modal Hashing (CMH) methods over unsupervised ones. Nevertheless, most existing supervised CMH methods still suffer from some limitations: 1) the observed labels are assumed to be complete and accurate, which may be impractical due to the missing and wrong class assignments, and 2) the semantic information is not fully excavated, especially for the semantic correlations among labels This paper proposes a Weakly-supervised enhAnced Semantic-aware Hashing (WASH) method which simultaneously estimates the label noises and performs enhanced semantic-aware hash learning. WASH employs the low-rank and sparse decomposition to alleviate the label noises, and a high-level semantic factor as well as a label correlation matrix is obtained by low-rank factorization on the noise-reduced labels. The low-rank semantic factors and multi-modal features are jointly factorized into a common subspace to reduce the heterogeneity gaps, which enhances the semantic awareness of shared representation. In this way, the hash codes can be obtained by binarizing the shared representation with pairwise semantic similarity preserved. Extensive experiments demonstrate that the proposed WASH method outperforms the state-of-the-art CMH methods.