MTFH: A Matrix Tri-Factorization Hashing Framework for Efficient Cross-Modal Retrieval

Hashing has recently sparked a great revolution in cross-modal retrieval because of its low storage cost and high query speed. Recent hashing methods often learn unified or equal-length hash codes to represent the multi-modal data make them intuitively comparable. However, such representations could inherently sacrifice their representation scalability from different modalities may not have one-to-one correspondence be encoded more efficiently by unequal lengths. To mitigate these problems, this paper exploits related relatively unexplored problem: encode heterogeneous with varying lengths generalize various challenging scenarios. end, generalized flexible framework, termed Matrix Tri-Factorization (MTFH), is proposed work seamlessly settings including paired unpaired data, equal length encoding More specifically, MTFH an efficient objective function flexibly modality-specific settings, while synchronously learning two semantic correlation matrices semantically correlate for As result, derived are meaningful tasks. Extensive experiments evaluated on public benchmark datasets highlight superiority under scenarios show competitive performance state-of-the-arts.