Bi-CMR: Bidirectional Reinforcement Guided Hashing for Effective Cross-Modal Retrieval

Cross-modal hashing has attracted considerable attention for large-scale multimodal data. Recent supervised cross-modal methods using multi-label networks utilize the semantics of multi-labels to enhance retrieval accuracy, where label hash codes are learned independently. However, all these assume that annotations reliably reflect relevance between their corresponding instances, which is not true in real applications. In this paper, we propose a novel framework called Bidirectional Reinforcement Guided Hashing Effective Cross-Modal Retrieval (Bi-CMR), exploits bidirectional learning relieve negative impact assumption. Specifically, forward procedure, highlight representative labels and learn reinforced by intra-modal semantic information, further adjust similarity matrix. backward adjusted matrix used guide matching instances. We construct two datasets with explicit instance pairs based on benchmark datasets. The Bi-CMR evaluated conducting extensive experiments over Experimental results prove superiority four state-of-the-art terms effectiveness.