Adaptive Label Correlation Based Asymmetric Discrete Hashing for Cross-Modal Retrieval

Hashing methods have captured much attention for cross-modal retrieval in recent years. Most existing approaches mainly focus on preserving the semantic similarity across heterogeneous modalities in a shared Hamming subspace, while the label information and potential correlations of multi-label semantics are not fully excavated. In this article, a novel Adaptive Label correlation based asymmEtric Cross-modal Hashing method, i.e., ALECH, is proposed for cross-modal retrieval. ALECH decomposes hash learning into two steps, hash codes learning and hash functions learning. For hash codes learning, the high-order semantic label correlations are adaptively exploited to guide the latent feature learning, while simultaneously generating the binary codes in a discrete manner. The asymmetric strategy is utilized to connect the latent feature space and Hamming space, and preserve the pairwise semantic similarity. Different from other two-step methods that directly adopt simple least-squares regression to learn hash functions based on binary codes, ALECH leverages both hash codes and semantic labels for hash functions learning which further preserves the similarity. Experiments on several benchmark datasets demonstrate that the proposed ALECH method outperforms the state-of-the-art cross-hashing methods.

Automated summary

Hashing methods have been widely studied for cross-modal retrieval. However, most existing approaches only focus on preserving semantic similarity and ignore the label information and multi-label correlations. In this article, a new method called Adaptive Label correlation based asymmEtric Cross-modal Hashing (ALECH) is proposed, which decomposes the hash learning into two steps: hash codes learning and hash functions learning. ALECH leverages adaptive label correlations and employs an asymmetric strategy to connect latent feature space and Hamming space to preserve semantic similarity. Experimental results on benchmark datasets demonstrate that ALECH outperforms state-of-the-art cross-hashing methods.