Cross-modal Correlation Learning Research Articles

Perceive, Reason, and Align: Context-guided cross-modal correlation learning for image–text retrieval

Due to the inconsistency in feature representations between different modalities, namely “Heterogeneous gap”, it remains a persistent challenge to correlate images and texts. Existing studies on image–text retrieval (ITR) mainly emphasize on inter-modal correlation learning through aligning instances or their patches from different modalities. However, it is hard to break through performance bottlenecks of ITR without powerfully supporting from intra-modal correlation. Unfortunately, few studies have sufficiently considered two critical tasks in intra-modal correlation learning: (1) intricate contextual information perceiving, and (2) intrinsic semantic relationships reasoning. Therefore, in this paper, we propose the Context-guided Cross-modal Correlation Learning (CCCL) framework for ITR under a novel paradigm: “Perceive, Reason, and Align”. Firstly, in the stage of “Perceive”, the context-guided mechanism based on the self-attention and gate mechanism is proposed to fully discover contextual information within modalities, eliminating unnecessary interactions between local-level patches. Secondly, in the stage of “Reason”, graph convolutional network with the residual structure is used to uncover relationships among patches within each modality to make reasonable inferences. Thirdly, in the stage of “Align”, to achieve precise inter-modal alignment, the complementarity between different modalities from both global-level and local-level is effectively mined and fused. Finally, to optimize our proposed CCCL framework, the hybrid loss is constructed by combining the cross-modal coherence term with the cross-modal alignment term. Our approach yields highly competitive results on two publicly available ITR datasets, that is, Flickr30K and MS-COCO.

Adaptive Adversarial Learning based cross-modal retrieval

There exists a heterogeneity gap between multi-modal data, hence it is difficult to directly measure the similarity between them. A common way to solve the problem is representation learning. Due to the unique adversarial optimization principle and the efficient cross-modal correlation learning ability, the cross-modal retrieval based on Generative Adversarial Network (GAN) has received significant attention recently. However, the potential semantic information is not fully explored in most GAN-based cross-modal learning approaches. In this paper, we propose a novel Adaptive Adversarial Learning (AAL) based cross-modal retrieval method. The generator of a specific modality projects heterogeneous data into the potential common subspace, while the discriminator is against the generator to maintain discriminability. In addition, three task-specific loss functions are designed in the generators to comprehensively exploit the semantic and label information. One problem is that directly optimizing the generator network will lead to ignoring the assessment of contribution to multi-loss functions. To overcome the above challenge, we present an adaptive balance strategy to match the appropriate contribution for each loss according to the degree of dispersion. Comprehensive experimental results on three widely-used databases show that the proposed method is effective and superior to the existing cross-modal retrieval methods.

Knowledge graph embedding by fusing multimodal content via cross-modal learning.

Knowledge graph embedding aims to learn representation vectors for the entities and relations. Most of the existing approaches learn the representation from the structural information in the triples, which neglects the content related to the entity and relation. Though there are some approaches proposed to exploit the related multimodal content to improve knowledge graph embedding, such as the text description and images associated with the entities, they are not effective to address the heterogeneity and cross-modal correlation constraint of different types of content and network structure. In this paper, we propose a multi-modal content fusion model (MMCF) for knowledge graph embedding. To effectively fuse the heterogenous data for knowledge graph embedding, such as text description, related images and structural information, a cross-modal correlation learning component is proposed. It first learns the intra-modal and inter-modal correlation to fuse the multimodal content of each entity, and then they are fused with the structure features by a gating network. Meanwhile, to enhance the features of relation, the features of the associated head entity and tail entity are fused to learn relation embedding. To effectively evaluate the proposed model, we compare it with other baselines in three datasets, i.e., FB-IMG, WN18RR and FB15k-237. Experiment result of link prediction demonstrates that our model outperforms the state-of-the-art in most of the metrics significantly, implying the superiority of the proposed method.

Deep Cross-Modal Face Naming for People News Retrieval

How to integrate multimodal information sources for face naming in multimodal news is a hot and yet challenging problem. A novel deep cross-modal face naming scheme is developed in this paper to facilitate more effective people news retrieval for large-scale multimodal news. This scheme integrates deep multimodal analysis, cross-modal correlation learning, and multimodal information mining, in which the efficient naming mechanism aims to cluster the deep features of different modalities into a common space to explore their inter-related correlations, and a special Web mining pattern is designed to optimize the name-face matching for rare non-celebrity. Such a cross-modal face naming model can be treated as a problem of bi-media semantic mapping and modeled as an inter-related correlation distribution over deep representations of multimodal news, in which the most important is to create more effective cross-modal name-face correlation and measure to what degree they are correlated. The experiments on a large number of public data from Yahoo! News have obtained very positive results and demonstrated the effectiveness of the proposed model.

CAESAR: concept augmentation based semantic representation for cross-modal retrieval

With the increasing amount of multimedia data, cross-modal retrieval has attracted more attentions in the area of multimedia and computer vision. To bridge the semantic gap between multi-modal data and improve the performance of retrieval, we propose an effective concept augmentation based method, named CAESAR, which is an end-to-end framework including cross-modal correlation learning and concept augmentation based semantic mapping learning. To enhance the representation and correlation learning, a novel multi-modal CNNs based CCA model is developed, which is to capture high-level semantic information during the cross-modal feature learning, and then capture maximal nonlinear correlation. In addition, to learn the semantic relationships between multi-modal samples, a concept learning model named CaeNet is proposed, which is realized by word2vec and LDA to capture the closer relations between texts and abstract concepts. Reenforce by the abstract concept information, cross-modal semantic mappings are learnt with a semantic alignment strategy. We conduct comprehensive experiments on four benchmark multimedia datasets. The results show that our method has great performance for cross-modal retrieval.

Deep cascaded cross-modal correlation learning for fine-grained sketch-based image retrieval

Fine-grained Sketch-based Image Retrieval (FG-SBIR), which utilizes hand-drawn sketches to search the target object images, has recently drawn much attention. It is a challenging task because sketches and images belong to different modalities and sketches are highly abstract and ambiguous. Existing solutions to this problem either focus on visual comparisons between sketches and images and ignore the multimodal characteristics of annotated images, or treat the retrieval as a one-time process. In this paper, we formulate FG-SBIR as a coarse-to-fine process, and propose a Deep Cascaded Cross-modal Ranking Model (DCCRM) that can exploit all the beneficial multimodal information in sketches and annotated images and improve both the retrieval efficiency and the top-K ranked effectiveness. Our goal concentrates on constructing deep representations for sketches, images, and descriptions, and learning the optimized deep correlations across such different domains. Thus for a given query sketch, its relevant images with fine-grained instance-level similarities in a specific category can be returned, and the strict requirement of the instance-level retrieval for FG-SBIR is satisfied. Very positive results have been obtained in our experiments by using a large quantity of public data.

Show and Tell in the Loop: Cross-Modal Circular Correlation Learning

Multimedia data with various modalities, such as image and text, are huge in quantity but have inconsistent distribution and representation. Many works have been done to break the boundary between image and text to measure their correlation. However, they focus on either the transformation to common subspace or the unidirectional generation from one to another individually, which cannot fully explore their interactions. It is noted that the bidirectional generation between image and text not only can provide complementary hints and mutually boost to learn cross-modal correlation but also cross-modal correlation learning can feed back to give comprehensive clues for promoting the cross-modal generation process. Therefore, we have the motivation that information transmission between image and text should be treated as a circular process, which aims to fully understand their latent correlation, and further realize cross-modal generation to produce both realistic images and text descriptions in a unified framework. In this paper, we propose the cross-modal circular correlation learning approach to perform both cross-modal correlation learning and generation simultaneously through an efficient circular learning training procedure. First, we propose the cross-modal circular learning model to perform an image-to-text caption and text-to-image synthesis circularly and learn common representation as a round-trip bridge, which can realize efficient interactions to fully exploit latent cross-modal correlations. Second, a unified bidirectional framework is proposed to conduct cross-modal mutual generation and is trained in an efficient circular process to enhance the generative ability of common representation, which can feed back circularly to further promote cross-modal correlation learning. In summary, we simultaneously perform cross-modal retrieval , image-to-text caption , and text-to-image synthesis in a unified framework with the circular learning process, which has high scalability and generality to realize universal cognition on the cross-modal data. We conduct extensive experiments to not only evaluate the correlation performance by cross-modal retrieval but also to show the generation effectiveness of both image caption and synthesis on the MS-COCO dataset.

Cross-modal learning for material perception using deep extreme learning machine

The material property of an object’s surface is critical for the tasks of robotic manipulation or interaction with its surrounding environment. Tactile sensing can provide rich information about the material characteristics of an object’s surface. Hence, it is important to convey and interpret tactile information of material properties to the users during interaction. In this paper, we propose a visual-tactile cross-modal retrieval framework to convey tactile information of surface material for perceptual estimation. In particular, we use tactile information of a new unknown surface material to retrieve perceptually similar surface from an available surface visual sample set. For the proposed framework, we develop a deep cross-modal correlation learning method, which incorporates the high-level nonlinear representation of deep extreme learning machine and class-paired correlation learning of cluster canonical correlation analysis. Experimental results on the publicly available dataset validate the effectiveness of the proposed framework and the method.

Deep Cross-Modal Correlation Learning for Audio and Lyrics in Music Retrieval

Deep cross-modal learning has successfully demonstrated excellent performance in cross-modal multimedia retrieval, with the aim of learning joint representations between different data modalities. Unfortunately, little research focuses on cross-modal correlation learning where temporal structures of different data modalities, such as audio and lyrics, should be taken into account. Stemming from the characteristic of temporal structures of music in nature, we are motivated to learn the deep sequential correlation between audio and lyrics. In this work, we propose a deep cross-modal correlation learning architecture involving two-branch deep neural networks for audio modality and text modality (lyrics). Data in different modalities are converted to the same canonical space where intermodal canonical correlation analysis is utilized as an objective function to calculate the similarity of temporal structures. This is the first study that uses deep architectures for learning the temporal correlation between audio and lyrics. A pretrained Doc2Vec model followed by fully connected layers is used to represent lyrics. Two significant contributions are made in the audio branch, as follows: (i) We propose an end-to-end network to learn cross-modal correlation between audio and lyrics, where feature extraction and correlation learning are simultaneously performed and joint representation is learned by considering temporal structures. (ii) And, as for feature extraction, we further represent an audio signal by a short sequence of local summaries (VGG16 features) and apply a recurrent neural network to compute a compact feature that better learns the temporal structures of music audio. Experimental results, using audio to retrieve lyrics or using lyrics to retrieve audio, verify the effectiveness of the proposed deep correlation learning architectures in cross-modal music retrieval.

CM-GANs

It is known that the inconsistent distributions and representations of different modalities, such as image and text, cause the heterogeneity gap, which makes it very challenging to correlate heterogeneous data and measure their similarities. Recently, generative adversarial networks (GANs) have been proposed and have shown their strong ability to model data distribution and learn discriminative representation. It has also been shown that adversarial learning can be fully exploited to learn discriminative common representations for bridging the heterogeneity gap. Inspired by this, we aim to effectively correlate large-scale heterogeneous data of different modalities with the power of GANs to model cross-modal joint distribution. In this article, we propose Cross-modal Generative Adversarial Networks (CM-GANs) with the following contributions. First, a cross-modal GAN architecture is proposed to model joint distribution over the data of different modalities. The inter-modality and intra-modality correlation can be explored simultaneously in generative and discriminative models. Both compete with each other to promote cross-modal correlation learning. Second, the cross-modal convolutional autoencoders with weight-sharing constraint are proposed to form the generative model. They not only exploit the cross-modal correlation for learning the common representations but also preserve reconstruction information for capturing the semantic consistency within each modality. Third, a cross-modal adversarial training mechanism is proposed, which uses two kinds of discriminative models to simultaneously conduct intra-modality and inter-modality discrimination. They can mutually boost to make the generated common representations more discriminative by the adversarial training process. In summary, our proposed CM-GAN approach can use GANs to perform cross-modal common representation learning by which the heterogeneous data can be effectively correlated. Extensive experiments are conducted to verify the performance of CM-GANs on cross-modal retrieval compared with 13 state-of-the-art methods on 4 cross-modal datasets.

Modality-specific Cross-modal Similarity Measurement with Recurrent Attention Network.

Nowadays, cross-modal retrieval plays an important role to flexibly find useful information across different modalities of data. Effectively measuring the similarity between different modalities of data is the key of cross-modal retrieval. Different modalities such as image and text have imbalanced and complementary relationship, and they contain unequal amount of information when describing the same semantics. For example, images often contain more details that cannot be demonstrated by textual descriptions and vice versa. Existing works based on Deep Neural Network (DNN) mostly construct one common space for different modalities, to find the latent alignments between them, which lose their exclusive modality-specific characteristics. Therefore, we propose modality-specific cross-modal similarity measurement (MCSM) approach by constructing the independent semantic space for each modality, which adopts an endto- end framework to directly generate modality-specific crossmodal similarity without explicit common representation. For each semantic space, modality-specific characteristics within one modality are fully exploited by recurrent attention network, while the data of another modality is projected into this space with attention based joint embedding, which utilizes the learned attention weights for guiding the fine-grained cross-modal correlation learning, and captures the imbalanced and complementary relationship between different modalities. Finally, the complementarity between the semantic spaces for different modalities is explored by adaptive fusion of the modality-specific cross-modal similarities to perform cross-modal retrieval. Experiments on the widely-used Wikipedia, Pascal Sentence, MS-COCO datasets as well as our constructed large-scale XMediaNet dataset verify the effectiveness of our proposed approach, outperforming 9 stateof- the-art methods.

CCL: Cross-modal Correlation Learning With Multigrained Fusion by Hierarchical Network

Cross-modal retrieval has become a highlighted research topic for retrieval across multimedia data such as image and text. A two-stage learning framework is widely adopted by most existing methods based on Deep Neural Network (DNN): The first learning stage is to generate separate representation for each modality, and the second learning stage is to get the cross-modal common representation. However, the existing methods have three limitations: (1) In the first learning stage, they only model intra-modality correlation, but ignore inter-modality correlation with rich complementary context. (2) In the second learning stage, they only adopt shallow networks with single-loss regularization, but ignore the intrinsic relevance of intra-modality and inter-modality correlation. (3) Only original instances are considered while the complementary fine-grained clues provided by their patches are ignored. For addressing the above problems, this paper proposes a cross-modal correlation learning (CCL) approach with multi-grained fusion by hierarchical network, and the contributions are as follows: (1) In the first learning stage, CCL exploits multi-level association with joint optimization to preserve the complementary context from intra-modality and inter-modality correlation simultaneously. (2) In the second learning stage, a multi-task learning strategy is designed to adaptively balance the intra-modality semantic category constraints and inter-modality pairwise similarity constraints. (3) CCL adopts multi-grained modeling, which fuses the coarse-grained instances and fine-grained patches to make cross-modal correlation more precise. Comparing with 13 state-of-the-art methods on 6 widely-used cross-modal datasets, the experimental results show our CCL approach achieves the best performance.

Corrections to “Cross-Modal Correlation Learning by Adaptive Hierarchical Semantic Aggregation” [Jun 16 1201-1216

Corrections to “Cross-Modal Correlation Learning by Adaptive Hierarchical Semantic Aggregation” [Jun 16 1201-1216

Cross-Modal Correlation Learning by Adaptive Hierarchical Semantic Aggregation

With the explosive growth of web data, effective and efficient technologies are in urgent need for retrieving semantically relevant contents of heterogeneous modalities. Previous studies devote efforts to modeling simple cross-modal statistical dependencies, and globally projecting the heterogeneous modalities into a measurable subspace. However, global projections cannot appropriately adapt to diverse contents, and the naturally existing multilevel semantic relation in web data is ignored. We study the problem of semantic coherent retrieval, where documents from different modalities should be ranked by the semantic relevance to the query. Accordingly, we propose TINA, a correlation learning method by adaptive hierarchical semantic aggregation. First, by joint modeling of content and ontology similarities, we build a semantic hierarchy to measure multilevel semantic relevance. Second, with a set of local linear projections and probabilistic membership functions, we propose two paradigms for local expert aggregation, i.e., local projection aggregation and local distance aggregation. To learn the cross-modal projections, we optimize the structure risk objective function that involves semantic coherence measurement, local projection consistency, and the complexity penalty of local projections. Compared to existing approaches, a better bias-variance tradeoff is achieved by TINA in real-world cross-modal correlation learning tasks. Extensive experiments on widely used NUS-WIDE and ICML-Challenge for image–text retrieval demonstrate that TINA better adapts to the multilevel semantic relation and content divergence, and, thus, outperforms state of the art with better semantic coherence.

Cluster-sensitive Structured Correlation Analysis for Web cross-modal retrieval

Modern cross-modal retrieving technology is required to find semantically relevant content from heterogeneous modalities. As previous studies construct unified dense correlation models on small scale cross-modal data, they are not capable of processing large scale Web data, because (a) the content of Web cross media is divergent; (b) the topic sensitive structure information in the high dimensional space is neglected; and (c) data should be organized as strictly corresponding pairs, which is not satisfied in real world scenarios. To address these challenges, we propose a cluster-sensitive cross-modal correlation learning framework. First, a set of cluster-sensitive correlation sub-models are learned instead of a unified correlation model, which better fits the content divergence in different modalities. We impose structured sparsity regularization on the projection vectors to learn a set of interpretable structured sparse correlation sub-models. Second, to compensate for the correspondence missing, we take full advantage of both intra-modal affinity and inter-modal co-occurrence. The projected coordinates of adjacent data within a modality tend to be similar, and the inconsistency of cluster-sensitive projection is minimized. The learned correlation model adapts to the content divergence and thus achieves better model generality and bias–variance trade-off. Extensive experiments on two large scale cross-modal data demonstrate the effectiveness of our approach.