Scalable Discriminative Discrete Hashing For Large-Scale Cross-Modal Retrieval

Abstract

Cross-modal hashing has received increasing research attentions due to its less storage and efficient retrieval. However, most existing cross-modal hashing methods focus only on exploring multi-modal information, while underestimate the significance of local and Euclidean structure information on the hashing learning procedure. In this paper, we propose a supervised discrete-based cross-modal hashing method, named Scalable Discriminative Discrete Hashing (SDDH), for cross-modal retrieval, where 1) the discrete hash codes are directly obtained by multi-modal features and semantic labels so that the quantization errors are dramatically reduced, and 2) the discrete hash codes simultaneously preserve the heterogeneous similarity and manifold information in the original space by employing matrix factoring with orthogonal and balanced constraints. Moreover, an efficient optimization is introduced to tackle the discrete solution, which makes the SDDH scalable to large-scale cross-modal retrieval. Empirical results on three widely-used benchmark databases clearly demonstrate the effectiveness and efficiency of the proposed method in comparison with state-of-the-arts.

Hybrid-attention based Feature-reconstructive Adversarial Hashing Networks for Cross-modal Retrieval

“…Asymmetric Correlation Quantization Hashing (ACQH) (Wang et al,2020) uses pairwise semantic similarity preservation and point by point label regression to construct combined quantization to generate hash codes. Scalable Discriminative Discrete Hashing (SDDH) (Qin et al,2021) introduces a composite learning framework for compact hash code learning. Most of the above cross-modal retrieval means based on hash learning use superficial structures to realize the feature extraction, which cannot explain the complex non-linear relationship between text and image.…”

Unsupervised Deep Cross-Modal Hashing Retrieval with Multiple Dense Networks

“…Hash technology aims to project data from any high-dimensional space into the binary space, and preserve the similarity relationship of data in the original space in the binary space. Since data is represented by compact binary codes, storage and search costs will be significantly reduced [11,12]. Because of these main advantages, hash-based methods have attracted wide attention, and hash learning has been widely used in crossmodal retrieval.…”