Cross-modal retrieval is a promising technique nowadays to find semantically similar instances in other modalities while a query instance is given from one modality. However, there still exists many challenges for reducing heterogeneous modality gap by embedding label information to discrete hash codes effectively, solving the binary optimization when generating unified hash codes and reducing the discrepancy of data distribution efficiently during common space learning. In order to overcome the above-mentioned challenges, we propose a Collaboratively Semantic alignment and Metric learning for cross-modal Hashing (CSMH) in this paper. Specifically, by a kernelization operation, CSMH first extracts the non-linear data features for each modality, which are projected into a latent subspace to align both marginal and conditional distributions simultaneously. Then, a maximum mean discrepancy-based metric strategy is customized to mitigate the distribution discrepancies among features from different modalities. Finally, semantic information obtained from the label similarity matrix, is further incorporated to embed the latent semantic structure into the discriminant subspace. Experimental results of CSMH and baseline methods on four widely-used datasets show that CSMH outperforms some state-of-the-art hashing baseline methods for cross-modal retrieval on efficiency and precision.

High-Order Graph (HOG) clustering has received much attention for its advantage of exploiting the rich intrinsic structure of data. However, the construction of HOG involves the generation of a large number of redundant walks, which dilutes the useful walks and thus leads to untrustworthy high-order similarity and, consequently, suboptimal clustering results may be obtained. We formalize this issue as the Weight Explosion (WE) problem. Furthermore, current works rarely focus on exploiting the correlation between multi-order graphs that can capture high-order relations at various levels. In this paper, we first analyze the pattern of redundant walks, also termed as noise, and subsequently propose a novel$h$-length Simple Path Search ($h$-SPS) algorithm to solve the WE problem.$h$-SPS aims to find valid walks to denoise HOG and thus avoids enumerating walks to report the similarity. Regarding the second problem, we propose a multi-order graphs fusion method, which adaptively integrates graphs of varying orders by solving a convex problem. This allows us to capture information across different order levels effectively. Extensive experiments on benchmark datasets demonstrate that our method11https://github.com/YonghaoChen511/DenoHOG can effectively solve the proposed WE problem, while also well exploiting the correlation of multi-order graphs.

Conventional hash-based retrieval method rely on the assumption that the query and database are of the identical domain. However, cross-domain problem often occurs in real-world applications, leading to the unsatisfactory performance of existing hashing methods. Recently, some researchers have put forward domain adaptation retrieval (DAR) under the perspective of domain adaptation (DA) and achieved promising results. But the following limitations still exist: 1) a single function is used to handle two challenges, i.e., domain adaptation and hashing, which is not flexible to explore enough underlying information for simultaneously accomplishing these challenges well; 2) non-dominant features in the sample are ignored; 3) the dissimilarity structure of dissimilar samples is not taken into account. To address the above problems, we propose a novel framework named two-step strategy (TSS) for domain adaptation retrieval, which advocates dividing DAR into two steps: DA step and hashing step. A DA function and a hash function are learned to handle the above two challenges, respectively, making the process more reasonable. Additionally, a discriminant semantic fusion loss is proposed to improve the discriminative ability among classes. Unlike other works that focus on discovering dominant features, we exploit the neglected non-dominant features and assign them attention with sinusoidal semantic embedding, actively creating a clear separation between classes. At last, we present an adaptive similarity preserving loss to preserve the similarity structure of the original data in all intra-domain and inter-domain hash codes. Extensive experiments on various datasets demonstrate that the proposed TSS achieves state-of-the-art performance.

Hashing-based techniques present appealing solutions for cross-modal retrieval due to its low storage requirements and excellent query efficiency. The majority of cross-modal hashing methods typically adopt equal-length encoding scheme to represent multimodal data and achieve cross-modal similarity search. However, such scheme can be regarded as a relatively strict limitation, because it sacrifices the flexible representation of multimodal data in reality and cannot always guarantee the optimal retrieval performance. To address the challenge, this paper focuses on encoding heterogeneous data with varying hash lengths. To achieve this purpose, we propose a flexible cross-modal hashing approach, named Two-stage Asymmetric Similarity Preserving Hashing, TASPH for short, which can be applied to both unequal-length and equal-length retrieval scenarios. Specifically, in the first stage, TASPH designs a novel discrete asymmetric strategy to learn the modality-specific hash codes with varying lengths, enabling a flexible representation of heterogeneous data. Simultaneously, TASPH utilizes two semantic transformation matrices to establish the semantic correlations between varying hash codes. Different from most of the existing approaches that employ relaxation solutions, TASPH satisfies the discrete constraints without any relaxation. In the second stage, the learned semantic transformation matrices are employed to alleviate cross-modal heterogeneity, which guarantees that TASPH can learn more powerful hash functions to improve the discriminative ability of hash codes. Abundant experiments conducted on three benchmark datasets demonstrate encouraging results compared with the state-of-the-art approaches under different retrieval scenarios.

In this paper, a novel Deep Semantic Space learning model with Intra-class Low-rank constraint (DSSIL) is proposed for cross-modal retrieval, which is composed of two subnetworks for modality-specific representation learning, followed by projection layers for common space mapping. In particular, DSSIL takes into account semantic consistency to fuse the cross-modal data in a high-level common space, and constrains the common representation matrix within the same class to be low-rank, in order to induce the intra-class representations more relevant. More formally, two regularization terms are devised for the two aspects, which have been incorporated into the objective of DSSIL. To optimize the modality-specific subnetworks and the projection layers simultaneously by exploiting the gradient decent directly, we approximate the nonconvex low-rank constraint by minimizing a few smallest singular values of the intra-class matrix with theoretical analysis. Extensive experiments conducted on three public datasets demonstrate the competitive superiority of DSSIL for cross-modal retrieval compared with the state-of-the-art methods.