Cross-media Correlation Research Articles

Cross-media correlation learning for web video event mining with integrated text semantics and network structural information

Cross-media correlation learning for web video event mining with integrated text semantics and network structural information

Unsupervised Visual–Textual Correlation Learning With Fine-Grained Semantic Alignment

With the rapid growth of multimedia data on the Internet, there has been a rapid rise in the demand for visual-textual cross-media retrieval between images and sentences. However, the heterogeneous property of visual and textual data brings huge challenges to measure the cross-media similarity for retrieval. Although existing methods have achieved great progress with the strong learning ability of the deep neural network, they rely heavily on the scale of training data with manual annotation, that is, either pairwise image-sentence annotation or category annotation as supervised information for visual-textual correlation learning, which are extremely labor and time consuming to collect. Without any pairwise or category annotation, it is highly challenging to construct a correlation between images and sentences due to their inconsistent distributions and representations. But people can naturally understand the correlation between visual and textual data in high-level semantic, and those images and sentences containing the same group of semantic concepts can be easily matched in human brain. Inspired by the above human cognitive process, this article proposes an unsupervised visual-textual correlation learning (UVCL) approach to construct correlations without any manual annotation. The contributions are summarized as follows: 1) unsupervised semantic-guided cross-media correlation mining is proposed to bridge the heterogeneous gap between visual and textual data. We measure the semantic matching degree between images and sentences, and generate descriptive sentences according to the concepts extracted from images to further augment the training data in an unsupervised manner. Therefore, the approach can exploit the semantic knowledge within both visual and textual data to reduce the gap between them for further correlation learning and 2) unsupervised visual-textual fine-grained semantic alignment is proposed to learn cross-media correlation by aligning the fine-grained visual local patches and textual keywords with fine-grained soft attention as well as semantic-guided hard attention, and the results can effectively highlight the fine-grained semantic information within both images and sentences to boost visual-textual alignment. Extensive experiments are conducted to perform visual-textual cross-media retrieval in unsupervised setting without any manual annotation on two widely used datasets, namely, Flickr-30K and MS-COCO, which verify the effectiveness of our proposed UVCL approach.

Cross-Media Semantic Correlation Learning Based on Deep Hash Network and Semantic Expansion for Social Network Cross-Media Search.

Cross-media search from large-scale social network big data has become increasingly valuable in our daily life because it can support querying different data modalities. Deep hash networks have shown high potential in achieving efficient and effective cross-media search performance. However, due to the fact that social network data often exhibit text sparsity, diversity, and noise characteristics, the search performance of existing methods often degrades when dealing with this data. In order to address this problem, this article proposes a novel end-to-end cross-media semantic correlation learning model based on a deep hash network and semantic expansion for social network cross-media search (DHNS). The approach combines deep network feature learning and hash-code quantization learning for multimodal data into a unified optimization architecture, which successfully preserves both intramedia similarity and intermedia correlation, by minimizing both cross-media correlation loss and binary hash quantization loss. In addition, our approach realizes semantic relationship expansion by constructing the image-word relation graph and mining the potential semantic relationship between images and words, and obtaining the semantic embedding based on both internal graph deep walk and an external knowledge base. Experimental results demonstrate that DHNS yields better cross-media search performance on standard benchmarks.

Unsupervised Cross-Media Retrieval Using Domain Adaptation With Scene Graph

Existing cross-media retrieval methods are usually conducted under the supervised setting, which need lots of annotated training data. Generally, it is extremely labor-consuming to annotate cross-media data. So unsupervised cross-media retrieval is highly demanded, which is very challenging because it has to handle heterogeneous distributions across different media types without any annotated information. To address the above challenge, this paper proposes Domain Adaptation with Scene Graph (DASG) approach, which transfers knowledge from the source domain to improve cross-media retrieval in the target domain. Our DASG approach takes Visual Genome as the source domain, which contains image knowledge in the form of scene graph. The main contributions of this paper are as follows: First, we propose to address unsupervised cross-media retrieval by domain adaptation. Instead of using the labor-consuming annotated information of cross-media data in the training stage, our DASG approach learns cross-media correlation knowledge from Visual Genome, and then transfers the knowledge to cross-media retrieval through media alignment and distribution alignment. Second, our DASG approach utilizes fine-grained information via scene graph representation to enhance generalization capability across domains. The generated scene graph representation builds (subject$\rightarrow $ relationship$\rightarrow $ object) triplets by exploiting objects and relationships within image and text, which makes the cross-media correlation more precise and promotes unsupervised cross-media retrieval. Third, we exploit the related tasks including object and relationship detection for learning more discriminative features across domains. Leveraging the semantic information of objects and relationships improves cross-media correlation learning for retrieval. Experiments on two widely-used cross-media retrieval datasets, namely Flickr-30K and MS-COCO, show the effectiveness of our DASG approach.

MAVA: Multi-level Adaptive Visual-textual Alignment by Cross-media Bi-attention Mechanism.

The rapidly developing information technology leads to a fast growth of visual and textual contents, and it comes with huge challenges to make correlation and perform crossmedia retrieval between images and sentences. Existing methods mainly explore cross-media correlation from either global-level instances as the whole images and sentences, or local-level fine-grained patches as the discriminative image regions and key words, which ignore the complementary information from the relation between local-level fine-grained patches. Naturally, relation understanding is highly important for learning crossmedia correlation. People focus on not only the alignment between discriminative image regions and key words, but also their relations lying in the visual and textual context. Therefore, in this paper, we propose Multi-level Adaptive Visual-textual Alignment (MAVA) approach with the following contributions. First, we propose cross-media multi-pathway fine-grained network to extract not only the local fine-grained patches as discriminative image regions and key words, but also visual relations between image regions as well as textual relations from the context of sentences, which contain complementary information to exploit fine-grained characteristics within different media types. Second, we propose visual-textual bi-attention mechanism to distinguish the fine-grained information with different saliency from both local and relation levels, which can provide more discriminative hints for correlation learning. Third, we propose cross-media multi-level adaptive alignment to explore global, local and relation alignments. An adaptive alignment strategy is further proposed to enhance the matched pairs of different media types, and discard those misalignments adaptively to learn more precise cross-media correlation. Extensive experiments are conducted to perform image-sentence matching on 2 widely-used cross-media datasets, namely Flickr-30K and MS-COCO, comparing with 10 state-of-the-art methods, which can fully verify the effectiveness of our proposed MAVA approach.

TPCKT: Two-Level Progressive Cross-Media Knowledge Transfer

As multimedia data have been the main form of big data, cross-media retrieval becomes a research hotspot. It provides a flexible retrieval paradigm across different media types, such as using an image query to retrieve the relevant text, video, and audio. An effective model to establish cross-media correlation is indispensable for retrieval. Existing methods usually rely on labeled data for model training, but it is extremely labor consuming to collect and label cross-media data. Under this situation, it is a key issue toward the real application to transfer knowledge from <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">existing data</i> to <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">new data</i> , for reducing the human labor. However, little attention has been paid to knowledge transfer between two <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">cross-media domains</i> . Therefore, this paper proposes the approach of two-level progressive cross-media knowledge transfer (TPCKT), which transfers knowledge from large-scale cross-media data, to boost the retrieval accuracy on cross-media data of another domain. Its contributions are: First, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">two-level adversarial transfer architecture</i> is proposed with domain discriminators in media-specific level and media-shared level, which have partially shared parameters to preserve cross-media consistency of transfer. The domain discrepancy between cross-media domains is fully reduced for boosting the retrieval accuracy. Second, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">progressive semantic transfer mechanism</i> is proposed to iteratively select semantically related categories in two cross-media domains for transfer. This drives the transfer process with ascending difficulties, for addressing the difficulty from different label spaces, and ensuring the robustness of transfer. For the experiment, the large-scale cross-media dataset PKU XMediaNet serves as the source domain, and three widely used small-scale datasets are adopted as the target domains to perform retrieval. Experimental results show the promising improvement gained by the proposed TPCKT.

Quintuple-Media Joint Correlation Learning With Deep Compression and Regularization

Multi-media data including image, video, text, audio, and 3D model, has been fast emerging on the Internet. Jointly correlating the data of various media types is a challenging task. With the considerable learning ability of deep network, existing works mainly construct multi-pathway network to learn cross-media correlation, where each pathway is for one media type. However, with number of media types increasing, existing methods face the problems of high repetition and complexity, leading to overfitting and poor generalization ability, which makes adverse effect on correlation learning. For addressing the above issues, we propose cross-media deep compression and regularization (CDCR) approach for quintuple-media joint correlation learning: 1) cross-media partial weight-sharing networks is proposed, where a part of parameters are commonly shared among multiple pathways, to exploit common characteristics across different media types for capturing intrinsic cross-media correlation; 2) we propose media-adaptive network pruning to drop connections between weakly-correlated neurons, which can emphasize media-specific characteristics adaptively; and 3) cross-media network regularization is proposed to utilize relationships among quintuple-media data, which can guarantee generalization ability and enhance intra-media and inter-media correlation. The experiments verify the effectiveness of our approach, which outperforms the state-of-the-art methods on two very challenging datasets, including a large-scale dataset PKU XMediaNet with more than 100 000 quintuple-media instances.

Reinforced Cross-Media Correlation Learning by Context-Aware Bidirectional Translation

The heterogeneity gap leads to inconsistent distributions and representations between image and text, which rises a challenging task to measure their similarities and construct cross-media correlation between them. The existing works mainly model the cross-media correlation in a common subspace, which causes insufficient correlation modeling in such third-party subspace with intermediate unidirectional transformation. Inspired by the recent advances of neural machine translation, which aims to establish a corresponding relationship between two entirely different languages, we can naturally discover that it has striking common characteristic with cross-media correlation learning to consider image and text as bilingual pairs, where the image is treated as a special kind of language to provide visual description, so that bidirectional transformation can be conducted between image and text to effectively explore cross-media correlation in the feature space of each media type. Thus, we propose a reinforced cross-media bidirectional translation (RCBT) approach to model the correlation between visual and textual descriptions. First, cross-media bidirectional translation mechanism is proposed to conduct direct transformation between the bilingual pairs of visual and textual descriptions bidirectionally, where the cross-media correlation can be effectively captured in both feature spaces of image and text through bidirectional translation training. Second, cross-media context-aware network with residual attention is proposed to exploit the rich spatial and temporal context hints with cross-media convolutional recurrent neural network, which can lead to more precise correlation learning for promoting bidirectional translation process. Third, cross-media reinforcement learning is proposed to perform a two-agent communication game played as a round between image and text to boost the bidirectional translation process, and we further extract inter-media and intra-media reward signals to provide complementary clues for learning cross-media correlation. Extensive experiments are conducted on cross-media retrieval to verify the effectiveness of our proposed RCBT approach, compared with 11 state-of-the-art methods on three cross-media datasets.

Cross-media analysis and reasoning: advances and directions

Cross-media analysis and reasoning is an active research area in computer science, and a promising direction for artificial intelligence. However, to the best of our knowledge, no existing work has summarized the state-of-the-art methods for cross-media analysis and reasoning or presented advances, challenges, and future directions for the field. To address these issues, we provide an overview as follows: (1) theory and model for cross-media uniform representation; (2) cross-media correlation understanding and deep mining; (3) cross-media knowledge graph construction and learning methodologies; (4) cross-media knowledge evolution and reasoning; (5) cross-media description and generation; (6) cross-media intelligent engines; and (7) cross-media intelligent applications. By presenting approaches, advances, and future directions in cross-media analysis and reasoning, our goal is not only to draw more attention to the state-of-the-art advances in the field, but also to provide technical insights by discussing the challenges and research directions in these areas.

Semi-Supervised Cross-Media Feature Learning With Unified Patch Graph Regularization

With the rapid growth of multimedia data such as text, image, video, audio, and 3-D model, cross-media retrieval has become increasingly important, because users can retrieve the results with various types of media by submitting a query of any media type. Comparing with single-media retrieval such as image retrieval and text retrieval, cross-media retrieval is better because it provides the retrieval results with all kinds of media at the same time. In this paper, we focus on how to learn cross-media features for different media types, which is a key challenge for cross-media retrieval. Existing methods either model different media types separately or only exploit the labeled multimedia data. Actually, the data from different media types with the same semantic category are complementary to each other, and jointly modeling them is able to improve the accuracy of cross-media retrieval. In addition, although the labeled data are accurate, they require a lot of human labor and thus are very scarce. To address the above problems, we propose a semi-supervised cross-media feature learning algorithm with unified patch graph regularization (S $^{\rm 2}$ UPG). Our motivation and contribution mainly lie in the following three aspects. First, existing methods only model different media types in different graphs, while we employ one joint graph to simultaneously model all the media types. The joint graph is able to fully exploit the semantic correlations among various media types, which are complementary to provide the rich hint for cross-media correlation. Second, existing methods only consider the original media instances (such as images, videos, texts, audios, and 3-D models) but ignore their patches, while we make full use of both the media instances and their patches in one graph. Cross-media patches could emphasize the important parts and make cross-media correlations more precise. Third, traditional semi-supervised learning methods only exploit single-media unlabeled instances, while our approach fully exploits cross-media unlabeled instances and their patches, which can increase the diversity of training data and boost the accuracy of cross-media retrieval. Comparing with the current state-of-the-art methods on three datasets, including the challenging XMedia dataset with five media types, the comprehensive experimental results show that our proposed approach performs better.

Multiple kernel visual-auditory representation learning for retrieval

Cross-media data representation, which focuses on semantics understanding of multimedia data in different modalities, is a rising hot topic in web media data analysis. The most challenging issues for cross-media data representation include: how to find underlying content-level data correlations and how to use such correlations in the representation model. Most traditional web media data analysis works are based on single modality data sources, such as Flickr images or YouTube videos, leaving cross-media data representation and semantics understanding wide open. In this paper, we propose a multiple kernel visual-auditory representation learning approach, which learns cross-media correlations from visual and auditory feature spaces with multiple kernel strategies. Besides, we give cross-media distance measure for image-audio retrieval in the mutual subspace of co-occurrence. Experiment results on the collected image-audio database are encouraging, and show that the performance of our approach is effective from multiple perspectives.

Content-oriented multimedia document understanding through cross-media correlation

This paper presents a novel method for multimedia document content analysis through modeling multimodal data correlations. We hypothesize that the correlation of different modalities from the same data source can help achieve better multimedia content understanding results than one which explores a single modality. We turn this task into two parts: multimedia data fusion and multimodal correlation propagation. During the first stage, we re-organize the training multimedia data into Modality semAntic Documents (MADs) after extracting quantized multimodal features, and then use multivariate Gaussian distributions to characterize the continuous quantity by latent topic modeling. Model parameters are asymmetrically learned to initialize multimodal correlations in the latent topic space. Accordingly, during the second stage, we construct a Multimodal Correlation Network (MCN) based on the initialized multimodal correlations, and a new mechanism of propagating inter-modality correlations and intra-modality similarities in MCN is further proposed to take the complementary from cross-modalities to facilitate multimedia content analysis. The experimental results of image-audio data retrieval on a 10-categories dataset and content-oriented web page recommendation on the USTODAY dataset show the effectiveness of our method.

Learning Cross-Media Joint Representation With Sparse and Semisupervised Regularization

Cross-media retrieval has become a key problem in both research and application, in which users can search results across all of the media types (text, image, audio, video, and 3-D) by submitting a query of any media type. How to measure the content similarity among different media is the key challenge. Existing cross-media retrieval methods usually focus on modeling the pairwise correlation or semantic information separately. In fact, these two kinds of information are complementary to each other and optimizing them simultaneously can further improve the accuracy. In this paper, we propose a novel feature learning algorithm for cross-media data, called joint representation learning (JRL), which is able to explore jointly the correlation and semantic information in a unified optimization framework. JRL integrates the sparse and semisupervised regularization for different media types into one unified optimization problem, while existing feature learning methods generally focus on a single media type. On one hand, JRL learns sparse projection matrix for different media simultaneously, so different media can align with each other, which is robust to the noise. On the other hand, both the labeled data and unlabeled data of different media types are explored. Unlabeled examples of different media types increase the diversity of training data and boost the performance of joint representation learning. Furthermore, JRL can not only reduce the dimension of the original features, but also incorporate the cross-media correlation into the final representation, which further improves the performance of both cross-media retrieval and single-media retrieval. Experiments on two datasets with up to five media types show the effectiveness of our proposed approach, as compared with the state-of-the-art methods.

Cross-media retrieval by intra-media and inter-media correlation mining

With the rapid development of multimedia content on the Internet, cross-media retrieval has become a key problem in both research and application. Cross-media retrieval is able to retrieve the results of the same semantics with the query, but with different media types. For instance, given a query image of Moraine Lake, besides retrieving the images about Moraine Lake, cross-media retrieval system can also retrieve the related media contents of different media types such as text description. As a result, measuring content similarity between different media is a challenging problem. In this paper, we propose a novel cross-media similarity measure. It considers both intra-media and inter-media correlation, which are ignored by existing works. Intra-media correlation focuses on semantic category information within each media, while inter-media correlation focuses on positive and negative correlations between different media types. Both of them are very important and their adaptive fusion can complement each other. To mine the intra-media correlation, we propose a heterogeneous similarity measure with nearest neighbors (HSNN). The heterogeneous similarity is obtained by computing the probability for two media objects belonging to the same semantic category. To mine the inter-media correlation, we propose a cross-media correlation propagation (CMCP) approach to simultaneously deal with positive and negative correlation between media objects of different media types, while existing works focus solely on the positive correlation. Negative correlation is very important because it provides effective exclusive information. The correlations are modeled as must-link constraints and cannot-link constraints, respectively. Furthermore, our approach is able to propagate the correlation between heterogeneous modalities. Finally, both HSNN and CMCP are flexible, so that any traditional similarity measure could be incorporated. An effective ranking model is learned by further fusion of multiple similarity measures through AdaRank for cross-media retrieval. The experimental results on two datasets show the effectiveness of our proposed approach, compared with state-of-the-art methods.

Somebody helps me: Travel video scene detection using web-based context

We conduct video scene detection with the aids of web-based context, especially for travel videos captured by amateur photographers in journeys. Correlations between personal videos and predefined travel schedules, which are used to retrieve related data from general-purpose image/video search engines, are discovered. Because scene boundaries are clearly defined in travel schedules, we segment videos into scenes by checking the discovered cross-media correlation. To make different modalities comparable, keyframes extracted from videos and images retrieved from web are represented by visual word histograms, and the problem of correlation determination is then transformed as an approximate sequence matching problem. We prioritize different visual words according to statistics of retrieved data, and evaluate similarity between images based on the weighting scheme. To systematically determine scene boundaries after finding cross-media correlation, we introduce an energy minimization framework to jointly consider visual, temporal, and context information. Experimental results verify the effectiveness of the proposed idea, and show that it is promising to utilize cross-media correlation and web-based context in media analysis.

Understanding visual-auditory correlation from heterogeneous features for cross-media retrieval

Cross-media retrieval is an interesting research topic, which seeks to remove the barriers among different modalities. To enable cross-media retrieval, it is needed to find the correlation measures between heterogeneous low-level features and to judge the semantic similarity. This paper presents a novel approach to learn cross-media correlation between visual features and auditory features for image-audio retrieval. A semi-supervised correlation preserving mapping (SSCPM) method is described to construct the isomorphic SSCPM subspace where canonical correlations between the original visual and auditory features are further preserved. Subspace optimization algorithm is proposed to improve the local image cluster and audio cluster quality in an interactive way. A unique relevance feedback strategy is developed to update the knowledge of cross-media correlation by learning from user behaviors, so retrieval performance is enhanced in a progressive manner. Experimental results show that the performance of our approach is effective.

Mining Semantic Correlation of Heterogeneous Multimedia Data for Cross-Media Retrieval

Although multimedia objects such as images, audios and texts are of different modalities, there are a great amount of semantic correlations among them. In this paper, we propose a method of transductive learning to mine the semantic correlations among media objects of different modalities so that to achieve the cross-media retrieval. Cross-media retrieval is a new kind of searching technology by which the query examples and the returned results can be of different modalities, e.g., to query images by an example of audio. First, according to the media objects features and their co-existence information, we construct a uniform cross-media correlation graph, in which media objects of different modalities are represented uniformly. To perform the cross-media retrieval, a positive score is assigned to the query example; the score spreads along the graph and media objects of target modality or MMDs with the highest scores are returned. To boost the retrieval performance, we also propose different approaches of long-term and short-term relevance feedback to mine the information contained in the positive and negative examples.

Exploring many-to-one speech-to-text correlation for web-based language learning

This article investigates the correlations between multimedia objects (particularly speech and text) involved in language lectures in order to design an effective presentation mechanism for web-based learning. The cross-media correlations are classified into implicit relations (retrieved by computing) and explicit relations (recorded during the preprocessing stage). The implicit temporal correlation between speech and text is primarily to help to negotiate supplementary lecture navigations like tele-pointer movement, lips-sync movement, and content scrolling. We propose a speech-text alignment framework, using an iterative algorithm based on local alignment, to probe many-to-one temporal correlations, and not the one-to-one only. The proposed framework is a more practical method for analyzing general language lectures, and the algorithm's time complexity conforms to the best-possible computation cost, O(nm) , without introducing additional computation. In addition, we have shown the feasibility of creating vivid presentations by exploiting implicit relations and artificially simulating some explicit media. To facilitate the navigation of integrated multimedia documents, we develop several visualization techniques for describing media correlations, including guidelines for speech-text correlations, visible-automatic scrolling, and levels of detail of timeline, to provide intuitive and easy-to-use random access mechanisms. We evaluated the performance of the analysis method and human perceptions of the synchronized presentation. The overall performance of the analysis method is that about 99.5% of the words analyzed are of a temporal error within 0.5 sec and the subjective evaluation result shows that the synchronized presentation is highly acceptable to human beings.

Toward better retrieval and presentation by exploring cross-media correlations

This paper addresses explicit correlation and implicit correlation between various media streams in a composite multimedia document, the so-called navigated hypermedia document in our language learning system, in order to facilitate document retrieval and synchronized presentation. For replaying a recorded lecture in a form as close as possible to the original classroom experience, we devised a capturing mechanism to explicitly record all the lecturing media streams and relations between them, including instructor's voice, slide change of the HTML lectures, and various guiding actions (e.g., tele-pointers, pen strokes, document scrolling, keyword highlighting, and text annotations) on HTML-based slides. In addition, for more effective learning, we study three different aspects - temporal, spatial, and content relation - of the implicit correlations that are inherently hidden between the media involved. The implicit relations are discovered by three designed processes: the speech-text alignment process for temporally synchronized speech-text presentation, the automatic scrolling process for the viewing window's spatial synchronization, and the content dependency checking process to ensure consistency of the content processed and the relations involved. The experimental results show that exploring cross-media correlations is helpful for system development in document presentation and retrieving. Users are allowed to replay a vivid and learning-effective multimedia lecture and to access the desired part of the document very easily via cross-media indexing. Hence the results have been applied to the development of online multimedia language learning systems aimed at improving students' English and Chinese language capabilities.