Deep Collaborative Discrete Hashing with Semantic-Invariant Structure

TL;DR

This paper introduces DCDH, a dual-stream deep hashing framework that enhances semantic correlation exploration and linguistic context integration for improved image retrieval performance.

Contribution

It proposes a novel collaborative discrete hashing method that combines visual and semantic features in a discriminative shared space, incorporating label reconstruction and focal loss.

Findings

Outperforms existing deep hashing methods in retrieval tasks.

Effectively captures semantic correlations and linguistic context.

Achieves superior accuracy on benchmark datasets.

Abstract

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.