Mutual Learning for Hashing: Unlocking Strong Hash Functions from Weak Supervision
This work addresses a limitation in hashing methods for large-scale image retrieval, offering an incremental improvement by combining existing approaches.
The paper tackles the problem of balancing global and local similarity in deep hashing for image retrieval by proposing Mutual Learning for Hashing (MLH), a weak-to-strong framework that enhances a center-based branch with knowledge from a pairwise-based branch, resulting in consistent outperformance over state-of-the-art methods across multiple benchmarks.
Deep hashing has been widely adopted for large-scale image retrieval, with numerous strategies proposed to optimize hash function learning. Pairwise-based methods are effective in learning hash functions that preserve local similarity relationships, whereas center-based methods typically achieve superior performance by more effectively capturing global data distributions. However, the strength of center-based methods in modeling global structures often comes at the expense of underutilizing important local similarity information. To address this limitation, we propose Mutual Learning for Hashing (MLH), a novel weak-to-strong framework that enhances a center-based hashing branch by transferring knowledge from a weaker pairwise-based branch. MLH consists of two branches: a strong center-based branch and a weaker pairwise-based branch. Through an iterative mutual learning process, the center-based branch leverages local similarity cues learned by the pairwise-based branch. Furthermore, inspired by the mixture-of-experts paradigm, we introduce a novel mixture-of-hash-experts module that enables effective cross-branch interaction, further enhancing the performance of both branches. Extensive experiments demonstrate that MLH consistently outperforms state-of-the-art hashing methods across multiple benchmark datasets.