Deep Collaborative Discrete Hashing with Semantic-Invariant Structure
This work addresses performance issues in deep hashing for tasks like image retrieval, but it appears incremental as it builds on existing methods with specific improvements.
The paper tackles the problem of existing deep hashing methods not fully exploring semantic correlations and linguistic context, which leads to poor performance, by proposing Deep Collaborative Discrete Hashing (DCDH), a dual-stream framework that constructs a discriminative common discrete space, resulting in demonstrated superiority in experiments.
Existing deep hashing approaches fail to fully explore semantic correlations and neglect the effect of linguistic context on visual attention learning, leading to inferior performance. This paper proposes a dual-stream learning framework, dubbed Deep Collaborative Discrete Hashing (DCDH), which constructs a discriminative common discrete space by collaboratively incorporating the shared and individual semantics deduced from visual features and semantic labels. Specifically, the context-aware representations are generated by employing the outer product of visual embeddings and semantic encodings. Moreover, we reconstruct the labels and introduce the focal loss to take advantage of frequent and rare concepts. The common binary code space is built on the joint learning of the visual representations attended by language, the semantic-invariant structure construction and the label distribution correction. Extensive experiments demonstrate the superiority of our method.