Video camouflaged object detection (VCOD) aims to identify objects that seamlessly blend into their surroundings in video sequences. Traditional methods merely rely on visual cues to capture inter-frame motion that reveals camouflaged objects. However, the high similarity between camouflaged objects and their environments often renders pure reliance on visual cues unreliable. Additionally, random motions including camera shaking and abrupt scene transitions also inevitably bring noise into the identification process. To overcome these challenges, we propose a Motion Reasoning Chain Network (MRCNet), a novel cross-modal VCOD framework that emulates the human thought process when observing camouflaged objects, i.e., motion reasoning. Specifically, we introduce a generative sampling strategy grounded in multimodal large language models (MLLMs) to bridge the implicit knowledge space of MLLMs and the explicit representation space regarding the attributes of camouflaged objects, thereby enabling the effective establishment of the motion reasoning chain tailored for VCOD. This process provides semantic guidance for visual comprehension of camouflaged objects through motion and concept attribute reasoning. To improve the identification capability of camouflaged objects, we develop motion representation learning driven by the motion reasoning chain. It introduces hierarchical de-biased motion prototype learning to mitigate hallucinations of MLLMs, boosting the motion perception. To learn precise prompts for the visual foundation model, cross-modal prompt learning further incorporates the de-biased concept prototype into visual representations to enhance the visual comprehension of camouflaged objects. Extensive experiments across three datasets demonstrate that MRCNet achieves state-of-the-art results on both general metrics and spatiotemporal consistency metrics.

Based on the message-passing paradigm, there has been an amount of research proposing diverse and impressive feature propagation mechanisms to improve the performance of GNNs. However, less focus has been put on feature transformation, another major operation of the message-passing framework. In this paper, we first empirically investigate the performance of the feature transformation operation in several typical GNNs. Unexpectedly, we notice that GNNs do not completely free up the power of the inherent feature transformation operation. By this observation, we propose the Bi-directional Knowledge Transfer (BiKT), a plug-and-play approach to unleash the potential of the feature transformation operations without modifying the original architecture. Taking the feature transformation operation as a derived representation learning model that shares parameters with the original GNN, the direct prediction by this model provides a topological-agnostic knowledge feedback that can further instruct the learning of GNN and the feature transformations therein. On this basis, BiKT not only allows us to acquire knowledge from both the GNN and its derived model but also promotes each other by injecting the knowledge into the other. In addition, a theoretical analysis is further provided to demonstrate that BiKT improves the generalization bound of the GNNs from the perspective of domain adaptation. An extensive group of experiments on up to 7 datasets with 5 typical GNNs demonstrates that BiKT brings up to 0.5% - 4% performance gain over the original GNN, which means a boosted GNN is obtained. Meanwhile, the derived model also shows a powerful performance to compete with or even surpass the original GNN, enabling us to flexibly apply it independently to some other specific downstream tasks.

Federated learning is a promising bridge that connects machine learning methods and multi-central medical data. It trains models using the local data, and protects the privacy of data. There are many methods for federated learning to aggregate models, especially personalized methods, which show relatively excellent performance. However, most of them excessively pay attention to global and local information while ignoring the random components during aggregating. That limits their performance in metrics like accuracy, specificity, and sensitivity. We propose a method (denoted by FedDiv) to make a balance between these metrics. The basic idea is to extract centralized features meanwhile filtering random components, and conduct personalized aggregation. These centralized features draw encoders’ attention, which enhances the performance of personalized models in specificity and sensitivity. Besides, they contain more global and local information, which is advantageous for personalized aggregation. Meanwhile, our personalized method preserves the local information as far as possible during aggregating models. These local information are the critical factor for the personalized models to perform better in accuracy. Finally, we validate this method in three public and one private medical datasets. Comparing with 14 federated methods, our method achieves the best performance in metrics including accuracy, specificity, sensitivity, and F1 score.

These years have witnessed remarkable progress in modeling user behaviour from personalized online services, especially knowledge graph-based recommendation systems. Meanwhile, more studies are focusing on aspects beyond recommendation performance, since such an observational data-driven paradigm is posing threats to both users and society in terms of trustworthiness. In fact, existing problem-oriented solutions still face significant challenges, as almost all of them suffer from the generality limitations to improve their trustworthiness in a uniform fashion. To address these issues, we propose a plug-and-playDebiasing framework forKnowledgeGraph-basedRecommendationSystems, also known as DiKGRS. Specifically, the Knowledge-augmented Pseudo-Samples Generation (KPSG) method, a novel data augmentation perspective, is proposed to explore more auxiliary information beyond observational user behaviors. Furthermore, the Debiasing Value Networks (DVN), is also developed to evaluate the reliability of generated pseudo-samples by modeling both the item popularity and user demographic bias in the platform. Moreover, an adaptive weighting coordination module is performed to coordinate the proposed DiKGRS framework and its backbones. Experimental results on four real-world datasets from different online service personalization scenarios have illustrated that the proposed framework can significantly improve the trustworthiness of existing knowledge graph-based recommendation systems. The code has been released public available at:https://github.com/alipay/A-Knowledge-augmented-Method-DiKGRS.

Despite the encouraging achievements, the practical application of recommendation systems still faces two key issues. The first is how to better understand the multimodal real-time requests that are the more mainstream request behavior in industrial scenarios; the other is how to effectively capture users' dynamic needs that change with temporal and spatial conditions. The breakthroughs in text understanding and generation capabilities of Large Language Models (LLMs) have demonstrated their tremendous potential in precise recommendation systems, particularly through the enhancement of the understanding of user intent. To address these issues, we propose a novel Spatial-Temporal Multimodal LLM for generative recommendation. Specifically, on the basis of the behavior data constructed from Alipay, spatial-temporal knowledge-guided fine-tuning module is proposed to capture specific needs in user real-time requests. Furthermore, a preference discovery module is developed to learn user preferences in visual queries from multimodal request perspective. Meanwhile, a personalized recommendation module is designed to aggregate spatial-temporal knowledge and user preferences for generative recommendation. Experimental results on a real-world deployed generative recommendation task from the ‘Explore' scenario in Alipay have demonstrated the effectiveness of the proposed framework.

Federated graph learning (FGL), as a privacy-preserving paradigm for distributed graph data training, aims to resolve graph data isolation issues under the framework of federated learning (FL). Despite the significant efforts made by existing FGL methods, two key challenges are still not well addressed: 1) how to mitigate graph heterogeneity in clients arising from feature deviation and structural deviation and 2) how to devise a favorable aggregation mechanism to maximize the client's benefit from collaborative training with privacy preserving. To tackle these issues, we take a perspective of latent factor and propose a HarmoFGL framework by Harmonizing graph neural network (GNN) latent factors for Federated Graph Learning, achieving cross-client federated training by coordinating personalized aggregation and client-level representation in a symbiotic space. To alleviate feature deviation, an implicit feature crossing (IFC) approach is proposed through the disentanglement of higher order feature dependency into client-universal and client-specific interactions. As for the graph heterogeneity induced by structural deviation, we establish a cross-client symbiotic parameter space spanned by GNN latent factors, on which a client-level representation is derived to characterize the inherent properties of clients. On the server side, on the basis of client relevance-driven personalized parameter aggregation, graph Laplacian regularization on client-level representations is implemented for collaborative training. Experimental results on five public graph datasets and two medical datasets demonstrate the effectiveness of HarmoFGL.

Graph condensation (GC) aims to distill the original graph into a small-scale graph, mitigating redundancy and accelerating GNN training. However, conventional GC approaches heavily rely on rigid GNNs and task-specific supervision. Such a dependency severely restricts their reusability and generalization across various tasks and architectures. In this work, we revisit the goal of ideal GC from the perspective of GNN optimization consistency, and then a generalized GC optimization objective is derived, by which those traditional GC methods can be viewed nicely as special cases of this optimization paradigm. Based on this, \textbf{Pre}-trained \textbf{G}raph \textbf{C}ondensation (\textbf{PreGC}) via optimal transport is proposed to transcend the limitations of task- and architecture-dependent GC methods. Specifically, a hybrid-interval graph diffusion augmentation is presented to suppress the weak generalization ability of the condensed graph on particular architectures by enhancing the uncertainty of node states. Meanwhile, the matching between optimal graph transport plan and representation transport plan is tactfully established to maintain semantic consistencies across source graph and condensed graph spaces, thereby freeing graph condensation from task dependencies. To further facilitate the adaptation of condensed graphs to various downstream tasks, a traceable semantic harmonizer from source nodes to condensed nodes is proposed to bridge semantic associations through the optimized representation transport plan in pre-training. Extensive experiments verify the superiority and versatility of PreGC, demonstrating its task-independent nature and seamless compatibility with arbitrary GNNs.

Facing the challenges of sparsity and long tail in thousands of Mini-apps recommendation scenarios deployed on Alipay platform, there is a great need for a simple, effective, and easy-to-deploy industrial solution. To address this issue, we follow the strategy of “divide and conquer” and propose a crowd-based recommendation model by using D eterminantal P oint P rocesse s on C rowd-wise M ixture- o f- E xperts (DPPs-CMoE). Specifically, under the guidance of DPPs-based prototypical tags, the user profiling space is sequentially divided into multiple crowds, with each of them taking on a unique latent specificity; Meanwhile, by treating the modeling of crowd specificity as one of multiple tasks, a crowd-wise architecture is adopted to seamlessly unify the multiple expert networks from the overall user space and the gating network from each of independent crowd spaces. The effectiveness of the proposed method has been illustrated in the experimental results on a mini-apps recommendation scenario deployed in Alipay APPs.

The Segment Anything Model (SAM), a prompt-driven foundational model, has demonstrated remarkable performance in natural image segmentation. However, its application in video camouflaged object detection (VCOD) en-counters challenges, chiefly stemming from the overlooked temporal-spatial associations and the unreliability of user-provided prompts for camouflaged objects that are difficult to discern with the naked eye. To tackle the above issues, we endow SAM with keen eyes and propose the Temporal-spatial Prompt SAM (TSP-SAM), a novel approach tailored for VCOD via an ingenious prompted learning scheme. Firstly, motion-driven self-prompt learning is employed to capture the camouflaged object, thereby bypassing the need for user-provided prompts. With the detected subtle motion cues across consecutive video frames, the overall movement of the camouflaged object is captured for more precise spa-tial localization. Subsequently, to eliminate the prompt bias resulting from inter-frame discontinuities, the long-range consistency within the video sequences is taken into account to promote the robustness of the self-prompts. It is also injected into the encoder of SAM to enhance the representational capabilities. Extensive experimental results on two benchmarks demonstrate that the proposed TSP-SAM achieves a significant improvement over the state-of-the-art methods. With the mIoU metric increasing by 7.8% and 9.6%, TSP-SAM emerges as a groundbreaking step forward in the field of VCOD.

Multimodal electronic health record (EHR) data can offer a holistic assessment of a patient's health status, supporting various predictive healthcare tasks. Recently, several studies have embraced the multitask learning approach in the healthcare domain, exploiting the inherent correlations among clinical tasks to predict multiple outcomes simultaneously. However, existing methods necessitate samples to possess complete labels for all tasks, which places heavy demands on the data and restricts the flexibility of the model. Meanwhile, within a multitask framework with multimodal inputs, how to comprehensively consider the information disparity among modalities and among tasks still remains a challenging problem. To tackle these issues, a unified healthcare prediction model, also named by \textbf{FlexCare}, is proposed to flexibly accommodate incomplete multimodal inputs, promoting the adaption to multiple healthcare tasks. The proposed model breaks the conventional paradigm of parallel multitask prediction by decomposing it into a series of asynchronous single-task prediction. Specifically, a task-agnostic multimodal information extraction module is presented to capture decorrelated representations of diverse intra- and inter-modality patterns. Taking full account of the information disparities between different modalities and different tasks, we present a task-guided hierarchical multimodal fusion module that integrates the refined modality-level representations into an individual patient-level representation. Experimental results on multiple tasks from MIMIC-IV/MIMIC-CXR/MIMIC-NOTE datasets demonstrate the effectiveness of the proposed method. Additionally, further analysis underscores the feasibility and potential of employing such a multitask strategy in the healthcare domain. The source code is available at https://github.com/mhxu1998/FlexCare.

Graph neural networks (GNNs) have shown remarkable performance on homophilic graph data while being far less impressive when handling non-homophilic graph data due to the inherent low-pass filtering property of GNNs. In general, since real-world graphs are often complex mixtures of diverse subgraph patterns, learning a universal spectral filter on the graph from the global perspective as in most current works may still suffer from great difficulty in adapting to the variation of local patterns. On the basis of the theoretical analysis of local patterns, we rethink the existing spectral filtering methods and propose theNode-oriented spectralFiltering forGraphNeuralNetwork (namely NFGNN). By estimating the node-oriented spectral filter for each node, NFGNN is provided with the capability of precise local node positioning via the generalized translated operator, thus discriminating the variations of local homophily patterns adaptively. Meanwhile, the utilization of re-parameterization brings a good trade-off between global consistency and local sensibility for learning the node-oriented spectral filters. Furthermore, we theoretically analyze the localization property of NFGNN, demonstrating that the signal after adaptive filtering is still positioned around the corresponding node. Extensive experimental results demonstrate that the proposed NFGNN achieves more favorable performance.

Based on multi-omics data and drug information, predicting the response of cancer cell lines to drugs is a crucial area of research in modern oncology, as it can promote the development of personalized treatments. Despite the promising performance achieved by existing models, most of them overlook the variations among different omics and lack effective integration of multi-omics data. Moreover, the explicit modeling of cell line/drug attribute and cell line-drug association has not been thoroughly investigated in existing approaches. To address these issues, we propose RedCDR, a dual relation distillation model for cancer drug response (CDR) prediction. Specifically, a parallel dual-branch architecture is designed to enable both the independent learning and interactive fusion feasible for cell line/drug attribute and cell line-drug association information. To facilitate the adaptive interacting integration of multi-omics data, the proposed multi-omics encoder introduces the multiple similarity relations between cell lines and takes the importance of different omics data into account. To accomplish knowledge transfer from the two independent attribute and association branches to their fusion, a dual relation distillation mechanism consisting of representation distillation and prediction distillation is presented. Experiments conducted on the GDSC and CCLE datasets show that RedCDR outperforms previous state-of-the-art approaches in CDR prediction.

Due to the various appearance of the polyps and the tiny contrast between the polyp area and its surrounding background, accurate polyp segmentation has become a challenging task. To tackle this issue, we introduce a boundary-enhanced framework for polyp segmentation, called the Focused on Boundary Segmentation (FoBS) framework, that leverages multi-level collaboration among sample, feature, and optimization. It places greater emphasis on the polyp boundary to improve the accuracy of segmentation. Firstly, a boundary-aware mixup method is designed to improve the model’s awareness of the boundary. More importantly, we propose deformable laplacian-based feature refining to explicitly strengthen the representation ability of the boundary features. It employs a deformable Laplacian refinement function to capture discriminative information from a deformable perceptual field, thereby improving its ability to adapt to boundary variations. In addition, we introduce the self-adjusting refinement coefficient learning that enables adaptive control over the refinement strength at each location. Furthermore, we develop a location-sensitive compensation criterion that assigns more importance to the degraded feature after feature refinement during optimization. Extensive quantitative and qualitative experiments on four polyp benchmarks demonstrate the effectiveness of our method for automatic polyp segmentation. Our code is available at https://github.com/TFboys-lzz/ FoBS.

In recent years, semi-supervised graph learning with data augmentation (DA) has been the most commonly used and best-performing method to improve model robustness in sparse scenarios with few labeled samples. However, most existing DA methods are based on the homogeneous graph, but none are specific for the heterogeneous graph. Differing from the homogeneous graph, DA in the heterogeneous graph faces greater challenges: heterogeneity of information requires DA strategies to effectively handle heterogeneous relations, which considers the information contribution of different types of neighbors and edges to the target nodes. Furthermore, over-squashing of information is caused by the negative curvature formed by the non-uniformity distribution and the strong clustering in a complex graph. To address these challenges, this article presents a novel method named HG-MDA (Semi-Supervised Heterogeneous Graph Learning with Multi-Level Data Augmentation). For the problem of heterogeneity of information in DA, node and topology augmentation strategies are proposed for the characteristics of the heterogeneous graph. Additionally, meta-relation-based attention is applied as one of the indexes for selecting augmented nodes and edges. For the problem of over-squashing of information, triangle-based edge adding and removing are designed to alleviate the negative curvature and bring the gain of topology. Finally, the loss function consists of the cross-entropy loss for labeled data and the consistency regularization for unlabeled data. To effectively fuse the prediction results of various DA strategies, sharpening is used. Existing experiments on public datasets (i.e., ACM, DBLP, and OGB) and the industry dataset MB show that HG-MDA outperforms current SOTA models. Additionally, HG-MDA is applied to user identification in internet finance scenarios, helping the business to add 30% key users, and increase loans and balances by 3.6%, 11.1%, and 9.8%.

Multi-task learning (MTL) is widely used in the online recommendation and financial services for multi-step conversion estimation, but current works often overlook the sequential dependence among tasks. In particular, sequential dependence multi-task learning (SDMTL) faces challenges in dealing with complex task correlations and extracting valuable information in real-world scenarios, leading to negative transfer and a deterioration in the performance. Herein, a systematic learning paradigm of the SDMTL problem is established for the first time, which applies to more general multi-step conversion scenarios with longer conversion paths or various task dependence relationships. Meanwhile, an SDMTL architecture, named Task-Aware Feature Extraction (TAFE), is designed to enable the dynamic task representation learning from a sample-wise view. TAFE selectively reconstructs the implicit shared information corresponding to each sample case and performs the explicit task-specific extraction under dependence constraints, which can avoid the negative transfer, resulting in more effective information sharing and joint representation learning. Extensive experiment results demonstrate the effectiveness and applicability of the proposed theoretical and implementation frameworks. Furthermore, the online evaluations at MYbank showed that TAFE had an average increase of 9.22% and 3.76% in various scenarios on the post-view click-through & conversion rate (CTCVR) estimation task. Currently, TAFE is deployed in an online platform to provide various traffic services.

Risk prediction, usually achieved by learning representations from patient’s physiological sequence or user’s behavioral sequence data, and has been widely applied in healthcare and finance. Despite that, some recent time-aware deep learning methods have led to superior performances in such sequence representation learning tasks, such improvement is limited due to a lack of guidance from hierarchical global view. To address this issue, we propose a novel end-to-end H ierarchical G lobal V iew-guided (HGV) sequence representation learning framework. Specifically, the Global Graph Embedding (GGE) module is proposed to learn sequential clip-aware representations from temporal correlation graph (TCG) at instance level. Furthermore, following the way of key-query attention, the harmonic β-attention (β-Attn) is also developed for making a global tradeoff between time-aware decay and observation significance at channel level adaptively. Moreover, the hierarchical representations at both instance level and channel level can be coordinated by the heterogeneous information aggregation under the guidance of global view. Experimental results on both healthcare risk prediction benchmark and SMEs credit overdue risk prediction task from the real-world industrial scenario in MYBank, Ant Group, have illustrated that the proposed model can achieve competitive prediction performance compared with other known baselines. The code has been released public available at: https://github.com/LiYouru0228/HGV.

While large enterprises are benefiting from their global supply chains in these years, it is not easy for Small and Medium-sized Enterprises (SMEs) to find supply chain partners. Treating it as a supply chain mining problem, some deep learning methods, especially knowledge graph (KG) enhanced ones, can achieve workable performance by utilizing explicit structure information from KG while considering effectiveness. However, such improvement is limited when facing the challenges of scalability, complexity, and noisiness in large-scale KGs. To address these issues, we propose a novelMeta-tagSupportedConnectivity representationLearning framework, also known as MSCL. Specifically, a Meta-tag Collaborative Filtering (MCF) method is proposed to highlight the representative schema from huge number of paths connecting two enterprises in large-scale KG. Furthermore, the DPPs-induced Hierarchical Path Sampling (DHPS), a novel sampling framework, is also developed to capture the latent connectivity pattern in KG more effectively. Moreover, the path-wise knowledge representations and the underlying information inherent in pairwise enterprises are aggregated by a connectivity representation learning (CRL) approach for SMEs supply chain mining. Experimental results from two real-world industries have illustrated that the proposed model can achieve competitive performance compared with other existing baselines.

Automatically recording surgical procedures and generating surgical reports are crucial for alleviating surgeons' workload and enabling them to concentrate more on the operations. Despite some achievements, there still exist several issues for the previous works: 1) failure to model the interactive relationship between surgical instruments and tissue; and 2) neglect of fine-grained differences within different surgical images in the same surgery. To address these two issues, we propose an improved scene graph-guided Transformer, also named by SGT++, to generate more accurate surgical report, in which the complex interactions between surgical instruments and tissue are learnt from both explicit and implicit perspectives. Specifically, to facilitate the understanding of the surgical scene graph under a graph learning framework, a simple yet effective approach is proposed for homogenizing the input heterogeneous scene graph. For the homogeneous scene graph that contains explicit structured and fine-grained semantic relationships, we design an attention-induced graph transformer for node aggregation via an explicit relation-aware encoder. In addition, to characterize the implicit relationships about the instrument, tissue, and the interaction between them, the implicit relational attention is proposed to take full advantage of the prior knowledge from the interactional prototype memory. With the learnt explicit and implicit relation-aware representations, they are then coalesced to obtain the fused relation-aware representations contributing to generating reports. Some comprehensive experiments on two surgical datasets show that the proposed STG++ model achieves state-of-the-art results.

Video camouflaged object detection aims to identify objects that are visually concealed within the surroundings in a video. Most of the existing methods fall into analyzing the implicit inter-frame motion to capture the camouflaged object. However, due to a lack of exploring the prior explicit motion of the camouflaged object, these works generally encounter difficulty in capturing the complete camouflaged object. To address this issue, we propose to integrate implicit and explicit motion learning into a unified framework, namelyImplicit-Explicit Motion Learning network (IMEX), for video camouflaged object detection. Specifically, to promote the identifiability of the camouflaged object, a cross-scale representation fusion was proposed for global inter-frame alignment. By establishing cross-scale temporal-spatial association and aggregating the temporal-spatial attentive representations, it also achieves an elimination of the implicit motion of inter-frame to some extent. Moreover, to further improve the discriminability of boundary regions of the detected object, an explicit motion-induced consistency preserving of camouflaged objects is proposed, in which the prior boundary-aware explicit motion field is leveraged to supervise the consistency of camouflaged objects in consecutive frames. Extensive experiments show that our proposed IMEX achieves substantial performance improvements by a large margin.

Automated radiology report generation aims to generate accurate and radiologist-like descriptions for the patient's images, which can greatly relieve the workload of radiologists. However, due to the data bias and long report problems, medical report generation has been a challenging task. In this article, we propose aFlexibleMulti-viewParadigm (FMVP) for medical report generation in a novel observation-to-concept manner. It first makes some medical observations automatically or with the help of a radiologist on the patient's image to obtain patient-related priori knowledge, just as radiologists do in practice. Furthermore, to bridge the gap betweenpretrainandgenerationphases, the hierarchical alignment is proposed to jointly conduct the implicit alignment between region-tag and the explicit global alignment of the image-report pair. Finally, a compatible decoder towards decoding the fused multi-view knowledge is proposed to capture more complementary information for the report generation, which breaks the traditional entrenched decoding mechanism guided by visual information. Extensive quantitative and qualitative experiments on the public MIMIC-CXR and IU-Xray datasets show that our model achieves competitive performance compared to state-of-the-art methods.

Zero-shot learning (ZSL) has received extensive attention recently especially in areas of fine-grained object recognition, retrieval, and image captioning. Due to the complete lack of training samples and high requirement of defense transferability, the ZSL model learned is particularly vulnerable against adversarial attacks. Recent work also showed adversarially robust generalization requires more data. This may significantly affect the robustness of ZSL. However, very few efforts have been devoted towards this direction. In this paper, we take an initial attempt, and propose a generic formulation to provide a systematical solution (namedATZSL) for learning a defensive ZSL model. It is capable of achieving better generalization on various adversarial objects recognition while only losing a negligible performance on clean images for unseen classes, by casting ZSL into a min-max optimization problem. To address it, we design a defensive relation prediction network, which can bridge the seen and unseen class domains via attributes to generalize prediction and defense strategy. Additionally, our framework can be extended to deal with the poisoned scenario of unseen class attributes. An extensive group of experiments are then presented, demonstrating that ATZSL obtains remarkably more favorable trade-off between model transferability and robustness, over currently available alternatives under various settings.

Recent years have witnessed the successful application of knowledge graph techniques in structured data processing, while how to incorporate knowledge from visual and textual modalities into knowledge graphs has been given less attention. To better organize them, Multimodal Knowledge Graphs (MKGs), comprising the structural triplets of traditional Knowledge Graphs (KGs) together with entity-related multimodal data (e.g., images and texts), have been introduced consecutively. However, it is still a great challenge to explore MKGs due to their inherent incompleteness. Although most existing Multimodal Knowledge Graph Completion (MKGC) approaches can infer missing triplets based on available factual triplets and multimodal information, they almost ignore the modal conflicts and supervisory effect, failing to achieve a more comprehensive understanding of entities. To address these issues, we propose a novel H ierarchical K nowledge A lignment ( HKA ) framework for MKGC. Specifically, a macro-knowledge alignment module is proposed to capture global semantic relevance between modalities for dealing with modal conflicts in MKG. Furthermore, a micro-knowledge alignment module is also developed to reveal the local consistency information through inter- and intra-modality supervisory effects more effectively. By integrating different modal predictions, a final decision can be made. Experimental results on three benchmark MKGC tasks have demonstrated the effectiveness of the proposed HKA framework.

Video summarization is an essential problem in signal processing, which intends to produce a concise summary of the original video. Existing video summarization approaches regard the task as a keyframe selection problem and generally construct the frame-wise representation by combining the long-range temporal dependency with either unimodal or bimodal information. The optimal keyframe should offer the semantic summarization of the whole content by exploiting the multimodal and shot-level hierarchical natures of videos, however, such natures are not fully exploited in existing methods. In this paper, we propose to construct a more powerful and robust frame-wise representation and predict the frame-level importance score in a fair and comprehensive manner. Specifically, we propose a multimodal hierarchical shot-aware convolutional network, denoted as MHSCNet, to enhance the frame-wise representation via combining the comprehensive available multimodal information. We further design a hierarchical ShotConv network to incorporate the adaptive shot-aware frame-level representation by considering the short-range and long-range temporal dependencies. Based on the learned shot-aware representations, MHSCNet can predict the frame-level importance score in the local and global view of the video. Extensive experiments on two standard video summarization datasets demonstrate that our proposed method consistently outperforms state-of-the-arts.

Multi-Target Cross Domain Recommendation(CDR) has attracted a surge of interest recently, which intends to improve the recommendation performance in multiple domains (or systems) simultaneously. Most existing multi-target CDR frameworks primarily rely on the existence of the majority of overlapped users across domains. However, general practical CDR scenarios cannot meet the strictly overlapping requirements and only share a small margin of common users across domains. Additionally, the majority of users have quite a few historical behaviors in such small-overlapping CDR scenarios. To tackle the aforementioned issues, we propose a simple-yet-effective neural node matching based framework for more general CDR settings, i.e., only (few) partially overlapped users exist across domains and most overlapped as well as non-overlapped users do have sparse interactions. The present framework mainly contains two modules: (i) intra-to-inter node matching module, and (ii) intra node complementing module. Concretely, the first module conducts intra-knowledge fusion within each domain and subsequent inter-knowledge fusion across domains by fully connected user-user homogeneous graph information aggregating. By doing this, the knowledge of all users, especially the non-overlapping users, could be well extracted and transferred without relying heavily on overlapping users. The second module introduces user-item matching to complement the potential missing interactions for each user and correct his/her under-represented representations, especially for the users with observed sparse interactions. Essentially, companion objectives are also inserted into each module to guide the knowledge transferring procedures, which leads to positive effects on multiple domains simultaneously. Extensive experiments on four multi-target CDR tasks from both public and real-world large-scale financial industry datasets demonstrate the remarkable performance of our proposed approach. Our code is publicly available at the link: https://github.com/WujiangXu/NMCDRR.

To effectively explore the supply chain relationships among Small and Medium-sized Enterprises (SMEs), some remarkable progress in such a relation modeling problem, especially knowledge graph-based methods have been witnessed during these years. As a typical link prediction task, supply chain prediction can usually predict the unknown future relationship facts between SMEs by utilizing the historical semantic connections between entities in knowledge graphs (KGs). However, it is still a great challenge for existing models as seldom of them can consider both temporal dependency and cooperative correlation of the connectivity pattern along the timeline synergistically. Accordingly, we propose a novel framework to learn joint relational co-evolution in Spatial-Temporal Knowledge Graphs (STKG). Specifically, on the base of the constructed large-scale financial STKG, a multi-view relational sequences mining method is proposed to reveal the semantic information from ontological concepts. Furthermore, a relational co-evolution learning module is also developed to capture the regularity of evolving connectivity patterns from the spatial-temporal view. Meanwhile, a multiple random subspace representation learning layer is also designed to improve both compatibility and complementarity during knowledge aggregation. Experimental results on large-scale SMEs supply chain prediction tasks from four real-world industries in China have illustrated the effectiveness of the proposed model.

In the remote sensing information interpretation tasks, compared with collecting lots of high-quality image labels for the target domain, a large amount of labeled remote sensing data from multiple source domains are generally available without any extra cost. In this paper, our work focuses on how to exploit the rich knowledge obtained from multiple source domains to guide the interpretation of the target scene, and we propose a novel framework called Collaborative Inter-domain and Intra-domain Alignments for multi-source domain adaptation, namely CoI2A, in which inter-domain and intra-domain alignments are well collaborated to reduce the distribution divergence across sources and target. To reduce the discrepancy across sources, the inter-source alignment is proposed to map multiple sources into a unified representation space. In addition, the cross-domain attention is introduced to enforce the intra-class compactness of the target. Inter-domain alignment aligns each source with target domain separately with the help of cross-domain attention. As for the intra-domain alignment, the multi-head attentive representations of the target obtained by cross-domain attention are correlated into a unified one. The experimental results obtained from different scene classification tasks demonstrate the superiority of our model.

Despite that path-based and embedding-based models with knowledge graphs (KGs) achieve better recommendation performance compared with other deep learning based methods, such improvement is limited due to a lack of modeling user's dynamic interest. To address this issue, we explore a principled model to provide semantic understanding of each item in user's historical interest sequence in KGs. Specifically, we propose a multi-granularity dynamic interest sequence learning method, which is based on knowledge-enhanced path mining and interest fluctuation signal discovery, to obtain semantic-enhanced paths. Furthermore, the paths are embedded by the SEP2Vec, and merged through the proposed entropy-aware pooling layer to obtain the user preference representation, which is then used to learn dynamic user interest sequence. Experimental results on two public datasets of movie and music recommendation, and two industrial datasets of personalized local service recommendation in Alipay App have illustrated that the proposed model can achieve significantly better prediction performance compared with other known baselines.

For an actual recommendation system, it generally involves a variety of heterogeneous interactive relationships, such as the typical user-user (U2U), item-item (I2I), and user-item (U2I) interaction relationships. With the application of graph neural networks (GNNs) in embedding various interactive relations, recommendation technology has made gratifying progress in recent years, which benefits lot from its powerful ability in relation modeling. However, most of the existing GNN-based methods fail to collaboratively explore the above heterogeneous multiple interactive relationships, including the internal correlations among multiple relationships and the intrinsic association behind different relationships. As a consequence, the user's personalized preference for the items to be recommended will not be well captured. In this paper, we propose aSylvester equation inducedCollaborativeRepresentationLearning framework (S-CRL) for recommendation system by utilizing the heterogeneous multiple interactive relationships. In particular, we ingeniously define a novel Sylvester equation to associate tactfully the multiple heterogeneous relations together. From the perspective of rating propagation, such Sylvester equation is shown theoretically to be the optimal solution of a local structure sensitive rating propagation function. Additionally, to seek more expressive embeddings about user and item, a layer-wise attention is introduced to aggregate the multi-hop information from U2U and I2I graphs, respectively, so as to promote the aggregation with the corresponding embeddings from the U2I interaction graph. Extensive experiments on three real-world datasets verify that our model achieves more favorable performance over currently representative methods.

BACKGROUND: Recently, with the foundation and development of gene ontology (GO) resources, numerous works have been proposed to compute functional similarity of genes and achieved series of successes in some research fields. Focusing on the calculation of the information content (IC) of terms is the main idea of these methods, which is essential for measuring functional similarity of genes. However, most approaches have some deficiencies, especially when measuring the IC of both GO terms and their corresponding annotated term sets. To this end, measuring functional similarity of genes accurately is still challenging. RESULTS: In this article, we proposed a novel gene functional similarity calculation method, which especially encapsulates the specificity of terms and edges (STE). The proposed method mainly contains three steps. Firstly, a novel computing model is put forward to compute the IC of terms. This model has the ability to exploit the specific structural information of GO terms. Secondly, the IC of term sets are computed by capturing the genetic structure between the terms contained in the set. Lastly, we measure the gene functional similarity according to the IC overlap ratio of the corresponding annotated genes sets. The proposed method accurately measures the IC of not only GO terms but also the annotated term sets by leveraging the specificity of edges in the GO graph. CONCLUSIONS: We conduct experiments on gene functional classification in biological pathways, gene expression datasets, and protein-protein interaction datasets. Extensive experimental results show the better performances of our proposed STE against several baseline methods.

Human identification is an important topic in event detection, person tracking, and public security. There have been numerous methods proposed for human identification, such as face identification, person re-identification, and gait identification. Typically, existing methods predominantly classify a queried image to a specific identity in an image gallery set (I2I). This is seriously limited for the scenario where only a textual description of the query or an attribute gallery set is available in a wide range of video surveillance applications (A2IorI2A). However, very few efforts have been devoted towards modality-free identification, i.e., identifying a query in a gallery set in a scalable way. In this work, we take an initial attempt, and formulate such a novelModality-FreeHumanIdentification (named MFHI) task as a generic zero-shot learning model in a scalable way. Meanwhile, it is capable of bridging the visual and semantic modalities by learning a discriminative prototype of each identity. In addition, the semantics-guided spatial attention is enforced on visual modality to obtain interpretable representations with both high global category-level and local attribute-level discrimination. Finally, we design and conduct an extensive group of experiments on two common challenging identification tasks, including face identification and person re-identification, demonstrating that our method outperforms a wide variety of state-of-the-art methods on modality-free human identification.

To bridge the gap between the source and target domains in unsupervised domain adaptation (UDA), the most common strategy puts focus on matching the marginal distributions in the feature space through adversarial learning. However, such category-agnostic global alignment lacks of exploiting the class-level joint distributions, causing the aligned distribution less discriminative. To address this issue, we propose in this paper a novel margin preserving self-paced contrastive Learning (MPSCL) model for cross-modal medical image segmentation. Unlike the conventional construction of contrastive pairs in contrastive learning, the domain-adaptive category prototypes are utilized to constitute the positive and negative sample pairs. With the guidance of progressively refined semantic prototypes, a novel margin preserving contrastive loss is proposed to boost the discriminability of embedded representation space. To enhance the supervision for contrastive learning, more informative pseudo-labels are generated in target domain in a self-paced way, thus benefiting the category-aware distribution alignment for UDA. Furthermore, the domain-invariant representations are learned through joint contrastive learning between the two domains. Extensive experiments on cross-modal cardiac segmentation tasks demonstrate that MPSCL significantly improves semantic segmentation performance, and outperforms a wide variety of state-of-the-art methods by a large margin.

Benefiting from the powerful expressive capability of graphs, graph-based approaches have been popularly applied to handle multi-modal medical data and achieved impressive performance in various biomedical applications. For disease prediction tasks, most existing graph-based methods tend to define the graph manually based on specified modality (e.g., demographic information), and then integrated other modalities to obtain the patient representation by Graph Representation Learning (GRL). However, constructing an appropriate graph in advance is not a simple matter for these methods. Meanwhile, the complex correlation between modalities is ignored. These factors inevitably yield the inadequacy of providing sufficient information about the patient's condition for a reliable diagnosis. To this end, we propose an end-to-end Multi-modal Graph Learning framework (MMGL) for disease prediction with multi-modality. To effectively exploit the rich information across multi-modality associated with the disease, modality-aware representation learning is proposed to aggregate the features of each modality by leveraging the correlation and complementarity between the modalities. Furthermore, instead of defining the graph manually, the latent graph structure is captured through an effective way of adaptive graph learning. It could be jointly optimized with the prediction model, thus revealing the intrinsic connections among samples. Our model is also applicable to the scenario of inductive learning for those unseen data. An extensive group of experiments on two disease prediction tasks demonstrates that the proposed MMGL achieves more favorable performance. The code of MMGL is available at https://github.com/SsGood/MMGL.

Due to the deteriorated conditions of illumination lack and uneven lighting, the performance of traditional background modeling methods is greatly limited for the surveillance of nighttime video. To make background modeling under nighttime scene performs as well as in daytime condition, we put forward a promising generation-based background modeling framework for foreground surveillance. With a pre-specified daytime reference image as background frame, the GAN based generation model, called N2DGAN, is trained to transfer each frame of nighttime video to a virtual daytime image with the same scene to the reference image except for the foreground part. Specifically, to balance the preservation of background scene and the foreground object(s) in generating the virtual daytime image, we presented a two-pathway generation model, in which the global and local sub-networks were well combined with spatial and temporal consistency constraints. For the sequence of generated virtual daytime images, a multi-scale Bayes model was further proposed to characterize pertinently the temporal variation of background. We manually labeled ground truth on the collected nightime video datasets for performance evaluation. The impressive results illustrated in both the main paper and supplementary show the effectiveness of our proposed approach.

In recent years, functional similarity has played an independent role in some biological fields such as gene clustering, gene functional prediction, and evaluation for protein-protein interaction. In this premise, some effective methods have already been proposed based on Gene Ontology (GO). Although these mainstream methods achieve the purpose for measuring gene functional similarity, they may have some deficiency when calculating the Information Content (IC) of GO terms. Consequently, measuring the functional similarity accurately is still a meaningful objective of research. In this paper, a novel method called SWE, is proposed for measuring gene functional similarity based on the GO graph. Firstly, an algorithm to measure terms' semantics based on their information in the GO graph is put forward. The information of GO terms mainly contains their depth, ancestors and descendants. Secondly, we calculate the IC of a term set by means of retrieving the inherited relationship between terms in a term set. Finally, the functional similarity between two genes is computed based on the IC overlap ratio of term sets annotating two genes respectively. Results demonstrate that SWE is superior to existing methods in some experiments such as functional classification of genes in a biological pathway, protein-protein interaction and gene expression experiment. Further analysis demonstrates that SWE takes not only the specificity of terms into account, but their information in the GO graph, both of which are shown to be consistent with human perspectives.

Graph representation learning aims to encode all nodes of a graph into low-dimensional vectors that will serve as input of many computer vision tasks. However, most existing algorithms ignore the existence of inherent data distribution and even noises. This may significantly increase the phenomenon of over-fitting and deteriorate the testing accuracy. In this paper, we propose a Distribution-induced Bidirectional Generative Adversarial Network (named DBGAN) for graph representation learning. Instead of the widely used Gaussian assumption, the prior distribution of latent representation in our DBGAN is estimated in a structure-aware way, which implicitly bridges the graph and content spaces by prototype learning. Thus discriminative and robust representations are generated for all nodes. Furthermore, to improve their generalization ability while preserving representation ability, the sample-level and distribution-level consistency are well balanced via a bidirectional adversarial learning framework. An extensive group of experiments is then carefully designed and presented, demonstrating that our DBGAN obtains remarkably more favorable trade-off between representation and robustness, and meanwhile is dimension-efficient, over currently available alternatives in various tasks.

For many machine learning algorithms, their success heavily depends on data representation. In this paper, we present an l2,1-norm constrained canonical correlation analysis (CCA) model, that is, L2,1-CCA, toward discovering compact and discriminative representation for the data associated with multiple views. To well exploit the complementary and coherent information across multiple views, the l2,1-norm is employed to constrain the canonical loadings and measure the canonical correlation loss term simultaneously. It enables, on the one hand, the canonical loadings to be with the capacity of variable selection for facilitating the interpretability of the learned canonical variables, and on the other hand, the learned canonical common representation keeps highly consistent with the most canonical variables from each view of the data. Meanwhile, the proposed L2,1-CCA can also be provided with the desired insensitivity to noise (outliers) to some degree. To solve the optimization problem, we develop an efficient alternating optimization algorithm and give its convergence analysis both theoretically and experimentally. Considerable experiment results on several real-world datasets have demonstrated that L2,1-CCA can achieve competitive or better performance in comparison with some representative approaches for multiview representation learning.

Existing enhancement methods are empirically expected to help the high-level end computer vision task: however, that is observed to not always be the case in practice. We focus on object or face detection in poor visibility enhancements caused by bad weathers (haze, rain) and low light conditions. To provide a more thorough examination and fair comparison, we introduce three benchmark sets collected in real-world hazy, rainy, and low-light conditions, respectively, with annotated objects/faces. We launched the UG2+ challenge Track 2 competition in IEEE CVPR 2019, aiming to evoke a comprehensive discussion and exploration about whether and how low-level vision techniques can benefit the high-level automatic visual recognition in various scenarios. To our best knowledge, this is the first and currently largest effort of its kind. Baseline results by cascading existing enhancement and detection models are reported, indicating the highly challenging nature of our new data as well as the large room for further technical innovations. Thanks to a large participation from the research community, we are able to analyze representative team solutions, striving to better identify the strengths and limitations of existing mindsets as well as the future directions.

Zero-Shot Learning (ZSL) has received extensive attention and successes in recent years especially in areas of fine-grained object recognition, retrieval, and image captioning. Key to ZSL is to transfer knowledge from the seen to the unseen classes via auxiliary semantic prototypes (e.g., word or attribute vectors). However, the popularly learned projection functions in previous works cannot generalize well due to non-visual components included in semantic prototypes. Besides, the incompleteness of provided prototypes and captured images has less been considered by the state-of-the-art approaches in ZSL. In this paper, we propose a hierarchical prototype learning formulation to provide a systematical solution (named HPL) for zero-shot recognition. Specifically, HPL is able to obtain discriminability on both seen and unseen class domains by learning visual prototypes respectively under the transductive setting. To narrow the gap of two domains, we further learn the interpretable super-prototypes in both visual and semantic spaces. Meanwhile, the two spaces are further bridged by maximizing their structural consistency. This not only facilitates the representativeness of visual prototypes, but also alleviates the loss of information of semantic prototypes. An extensive group of experiments are then carefully designed and presented, demonstrating that HPL obtains remarkably more favorable efficiency and effectiveness, over currently available alternatives under various settings.

Bike-sharing systems, aiming at meeting the public’s need for ”last mile” transportation, are becoming popular in recent years. With an accurate demand prediction model, shared bikes, though with a limited amount, can be effectively utilized whenever and wherever there are travel demands. Despite that some deep learning methods, especially long shortterm memory neural networks (LSTMs), can improve the performance of traditional demand prediction methods only based on temporal representation, such improvement is limited due to a lack of mining complex spatial-temporal relations. To address this issue, we proposed a novel model named STG2Vec to learn the representation from heterogeneous spatial-temporal graph. Specifically, we developed an event-flow serializing method to encode the evolution of dynamic heterogeneous graph into a special language pattern such as word sequence in a corpus. Furthermore, a dynamic attention-based graph embedding model is introduced to obtain an importance-awareness vectorized representation of the event flow. Additionally, together with other multi-source information such as geographical position, historical transition patterns and weather, e.g., the representation learned by STG2Vec can be fed into the LSTMs for temporal modeling. Experimental results from Citi-Bike electronic usage records dataset in New York City have illustrated that the proposed model can achieve competitive prediction performance compared with its variants and other baseline models.

Non-uniform blur, mainly caused by camera shake and motions of multiple objects, is one of the most common causes of image quality degradation. However, the traditional blind deblurring methods based on blur kernel estimation do not perform well on complicated non-uniform motion blurs. However, recent studies show that GAN-based approaches achieve impressive performance on deblurring tasks. In this letter, to further improve the performance of GAN-based methods on deblurring tasks, we propose an edge heuristic multi-scale generative adversarial network (GAN), which uses the coarse-to-fine scheme to restore clear images in an end-to-end manner. In particular, an edge-generated network is designed to generate sharp edges as auxiliary information to guide the deblurring process. Furthermore, We propose a hierarchical content loss function for deblurring tasks. Extensive experiments on different datasets show that our method achieves state-of-the-art performance in dynamic scene deblurring.

Prototype selection aims to remove redundancy and irrelevance from large-scale data by selecting an informative subset, which makes it possible to see all data from a few prototypes. However, due to the outliers and uncertain distribution of the data, the selected prototypes are generally less representative and diversified. To alleviate this issue, we develop, in this paper, a ℓ1-norm-induced discriminative prototype selection model (ℓ1-ProSe). Inspired by the good performance of sparse representation, the sparsity property of data is rationally exploited in the formulated model. Meanwhile, to characterize the pairwise similarity in the learned sparse representation space, a more promising ℓ1-norm metric is applied for robust selection instead of the popularly used Euclidean metric in previous works. Considering the convexity of the model to be solved, a composite block coordinate descent solver is presented with rigorous theoretical analysis on its convergence. Furthermore, we extend our model to support online prototype selection by using already obtained prototypes and newly arrived data. Experimental results on synthetic data sets and some applications such as video summarization, motion segmentation, and scene categorization demonstrate that the proposed method is considerably superior to the state-of-the-art methods in the prototype selection.

Prototype selection is a promising technique for removing redundancy and irrelevance from large-scale data. Here, we consider it as a task assignment problem, which refers to assigning each element of a source set to one representative, i.e., prototype. However, due to the outliers and uncertain distribution on source, the selected prototypes are generally less representative and interesting. To alleviate this issue, we develop in this paper a Self-supervised Deep Low-rank Assignment model (SDLA). By dynamically integrating a low-rank assignment model with deep representation learning, our model effectively ensures the goodness-of-exemplar and goodness-of-discrimination of selected prototypes. Specifically, on the basis of a denoising autoencoder, dissimilarity metrics on source are continuously self-refined in embedding space with weak supervision from selected prototypes, thus preserving categorical similarity. Conversely, working on this metric space, similar samples tend to select the same prototypes by designing a low-rank assignment model. Experimental results on applications like text clustering and image classification (using prototypes) demonstrate our method is considerably superior to the state-of-the-art methods in prototype selection.

Video analytics, also known as video content analysis, involves a variety of pivotal tasks, such as video segmentation and recognition. In essence, performing most of these tasks can be viewed as learning an assignment model. Here, assignment model refers to what each element in a target set is assigned to the element in an opposite source set under some kind of constraint. Existing assignment models generally suffer from some limitations. For instance, imperfect results can be obtained when the source set is artificially provided with less representativeness, or part of the target set has expected assignments, thus significantly limiting the application of assignment model. To alleviate these issues, we develop, in this paper, a general assignment model to dynamically learn a source set by minimizing structural dissimilarity in a low-dimensional space. Furthermore, by associating the expected assignment solution with empirical assignment indicator via a consistency boosting strategy, the proposed assignment model is provided with a potential powerful generalization ability to deal flexibly with the unsupervised, semi-supervised, and fully supervised scenarios in assignment model learning. Considering the separability of both objective and constraints to be solved, an alternating direction method of multipliers solver is presented with rigorous theoretical analysis on its convergence. Experimental results on video analysis, including motion segmentation, activities recognition, and scene categorization, demonstrate that the proposed general assignment model is considerable superior to the state-of-the-art methods.

A growing number of multi-view data arises naturally in many scenarios, including medical diagnosis, webpage classification, and multimedia analysis. A challenge in learning from multi-view data is that not all instances are fully represented in all views, resulting in missing view data. In this paper, we focus on feature-level completion for missing view of multi-view data. Aiming at capturing both semantic complementarity and identical distribution among different views, an Isomorphic Linear Correlation Analysis (ILCA) method is proposed to linearly map multi-view data to a feature-isomorphic subspace through learning a set of excellent isomorphic features, thereby unfolding the shared information from different views. Meanwhile, we assume that missing view obeys normal distribution. Then, the missing view data matrix can be modeled as a low-rank component plus a sparse contribution. Thus, to accomplish missing view completion, an Identical Distribution Pursuit Completion (IDPC) model based on the learned features is proposed, in which the identical distribution constraint of missing view to the other available one in the feature-isomorphic subspace is fully exploited. Comprehensive experiments on several multi-view datasets demonstrate that our proposed framework yields promising results.

Personalized recommendation aims to use the historical behavior of users to recommend new items that are likely to be of interest to them. Due to a tiny improvement of it can lead to a huge profits, lots of giant e-commerce companies, such as Amazon, Alibaba and eBay, have put their great effort on this field. In this paper, we formulate the problem of personalized recommendation as a tree based regression task and present a de-biased DART ensemble model. To generate compact personalized profile of user, a decision tree based sparse encoding approach is first proposed to refine those raw behaviors, like clicks, views, and purchases from user-item interaction. Considering the problem of class imbalance in training the predictive models, resampling based de-biased DART model is utilized to form multiple weak predictors for recommendation. Thus, based on these learnt weak predictors, a stronger predictor can be built. Experimental results on real commercial dataset demonstrate the effectiveness of the proposed personalized recommendation model.

Recently, convolutional neural network (CNN) visual features have demonstrated their powerful ability as a universal representation for various recognition tasks. In this paper, cross-modal retrieval with CNN visual features is implemented with several classic methods. Specifically, off-the-shelf CNN visual features are extracted from the CNN model, which is pretrained on ImageNet with more than one million images from 1000 object categories, as a generic image representation to tackle cross-modal retrieval. To further enhance the representational ability of CNN visual features, based on the pretrained CNN model on ImageNet, a fine-tuning step is performed by using the open source Caffe CNN library for each target data set. Besides, we propose a deep semantic matching method to address the cross-modal retrieval problem with respect to samples which are annotated with one or multiple labels. Extensive experiments on five popular publicly available data sets well demonstrate the superiority of CNN visual features for cross-modal retrieval.

In this article, we investigate the cross-media retrieval between images and text, that is, using image to search text (I2T) and using text to search images (T2I). Existing cross-media retrieval methods usually learn one couple of projections, by which the original features of images and text can be projected into a common latent space to measure the content similarity. However, using the same projections for the two different retrieval tasks (I2T and T2I) may lead to a tradeoff between their respective performances, rather than their best performances. Different from previous works, we propose a modality-dependent cross-media retrieval (MDCR) model, where two couples of projections are learned for different cross-media retrieval tasks instead of one couple of projections. Specifically, by jointly optimizing the correlation between images and text and the linear regression from one modal space (image or text) to the semantic space, two couples of mappings are learned to project images and text from their original feature spaces into two common latent subspaces (one for I2T and the other for T2I). Extensive experiments show the superiority of the proposed MDCR compared with other methods. In particular, based on the 4,096-dimensional convolutional neural network (CNN) visual feature and 100-dimensional Latent Dirichlet Allocation (LDA) textual feature, the mAP of the proposed method achieves the mAP score of 41.5%, which is a new state-of-the-art performance on the Wikipedia dataset.

Cross-media retrieval has received increasing interest in recent years, which aims to addressing the semantic correlation issues within rich media. As two key aspects, cross-media representation and indexing have been studied for dealing with cross-media similarity measure and the scalability issue, respectively. In this paper, we propose a new cross-media hashing scheme, called Centroid Approaching Cross-Media Hashing (CAMH), to handle both cross-media representation and indexing simultaneously. Different from existing indexing methods, the proposed method introduces semantic category information into the learning procedure, leading to more exact hash codes of multiple media type instances. In addition, we present a comparative study of cross-media indexing methods under a unique evaluation framework. Extensive experiments on two commonly used datasets demonstrate the good performance in terms of search accuracy and time complexity.

Existing recommender systems place emphasis on personalization to achieve promising accuracy. However, in the context of multiple domain, users are likely to seek the same behaviors as domain authorities. This conformity effect provides a wealth of prior knowledge when it comes to multi-domain recommendation, but has not been fully exploited. In particular, users whose behaviors are significant similar with the public tastes can be viewed as domain authorities. To detect these users meanwhile embed conformity into recommendation, a domain-specific similarity matrix is intuitively employed. Therefore, a collective similarity is obtained to leverage the conformity with personalization. In this paper, we establish a Collective Structure Sparse Representation(CSSR) method for multi-domain recommendation. Based on adaptive $k$-Nearest-Neighbor framework, we impose the lasso and group lasso penalties as well as least square loss to jointly optimize the collective similarity. Experimental results on real-world data confirm the effectiveness of the proposed method.

Traditional clustering and feature selection methods consider the data matrix as static. However, the data matrices evolve smoothly over time in many applications. A simple approach to learn from these time-evolving data matrices is to analyze them separately. Such strategy ignores the time-dependent nature of the underlying data. In this paper, we propose two formulations for evolutionary co-clustering and feature selection based on the fused Lasso regularization. The evolutionary co-clustering formulation is able to identify smoothly varying hidden block structures embedded into the matrices along the temporal dimension. Our formulation is very flexible and allows for imposing smoothness constraints over only one dimension of the data matrices. The evolutionary feature selection formulation can uncover shared features in clustering from time-evolving data matrices. We show that the optimization problems involved are non-convex, non-smooth and non-separable. To compute the solutions efficiently, we develop a two-step procedure that optimizes the objective function iteratively. We evaluate the proposed formulations using the Allen Developing Mouse Brain Atlas data. Results show that our formulations consistently outperform prior methods.

Independent component analysis with soft reconstruction cost (RICA) has been recently proposed to linearly learn sparse representation with an overcomplete basis, and this technique exhibits promising performance even on unwhitened data. However, linear RICA may not be effective for the majority of real-world data because nonlinearly separable data structure pervasively exists in original data space. Meanwhile, RICA is essentially an unsupervised method and does not employ class information. Motivated by the success of the kernel trick that maps a nonlinearly separable data structure into a linearly separable case in a high-dimensional feature space, we propose a kernel RICA (kRICA) model to nonlinearly capture sparse representation in feature space. Furthermore, we extend the unsupervised kRICA to a supervised one by introducing a class-driven discrimination constraint, such that the data samples from the same class are well represented on the basis of the corresponding subset of basis vectors. This discrimination constraint minimizes inhomogeneous representation energy and maximizes homogeneous representation energy simultaneously, which is essentially equivalent to maximizing between-class scatter and minimizing within-class scatter at the same time in an implicit manner. Experimental results demonstrate that the proposed algorithm is more effective than other state-of-the-art methods on several datasets.

In this paper, we propose a cross-media regularization framework to enhance image understanding which can benefit image retrieval, classification and so on. The goal of cross-media regularization is to find regularization projections by exploiting the correlations between visual features and textual features. Thus, the original noisy distribution of visual features can be refined by leveraging the discriminative distribution of the corresponding textual features. Within the proposed cross-media regularization framework, a mid-level representation is built by jointly projecting both visual and textual features into a shared feature subspace, which leads to transferring of the discriminative semantic characteristic embedded in the textual modality into the corresponding visual modality. Meanwhile, the discriminative characteristic of textual features can also be boosted simultaneously. The experimental results demonstrate that the proposed mid-level space learning process can remarkably improve the search quality and outperform the existing semantic regularization methods.

Nonnegative matrix factorization (NMF) is a useful technique to explore a parts-based representation by decomposing the original data matrix into a few parts-based basis vectors and encodings with nonnegative constraints. It has been widely used in image processing and pattern recognition tasks due to its psychological and physiological interpretation of natural data whose representation may be parts-based in human brain. However, the nonnegative constraint for matrix factorization is generally not sufficient to produce representations that are robust to local transformations. To overcome this problem, in this paper, we proposed a topographic NMF (TNMF), which imposes a topographic constraint on the encoding factor as a regularizer during matrix factorization. In essence, the topographic constraint is a two-layered network, which contains the square nonlinearity in the first layer and the square-root nonlinearity in the second layer. By pooling together the structure-correlated features belonging to the same hidden topic, the TNMF will force the encodings to be organized in a topographical map. Thus, the feature invariance can be promoted. Some experiments carried out on three standard datasets validate the effectiveness of our method in comparison to the state-of-the-art approaches.

Recently, some global contrast-based salient region detection models have been proposed based on only the low-level feature of color. It is necessary to consider both color and orientation features to overcome their limitations, and thus improve the performance of salient region detection for images with low-contrast in color and high-contrast in orientation. In addition, the existing fusion methods for different feature maps, like the simple averaging method and the selective method, are not effective sufficiently. To overcome these limitations of existing salient region detection models, we propose a novel salient region model based on the bottom-up and top-down mechanisms: the color contrast and orientation contrast are adopted to calculate the bottom-up feature maps, while the top-down cue of depth-from-focus from the same single image is used to guide the generation of final salient regions, since depth-from-focus reflects the photographer's preference and knowledge of the task. A more general and effective fusion method is designed to combine the bottom-up feature maps. According to the degree-of-scattering and eccentricities of feature maps, the proposed fusion method can assign adaptive weights to different feature maps to reflect the confidence level of each feature map. The depth-from-focus of the image as a significant top-down feature for visual attention in the image is used to guide the salient regions during the fusion process; with its aid, the proposed fusion method can filter out the background and highlight salient regions for the image. Experimental results show that the proposed model outperforms the state-of-the-art models on three public available data sets.

In some real world applications, like information retrieval and data classification, we often are confronted with the situation that the same semantic concept can be expressed using different views with similar information. Thus, how to obtain a certain Semantically Consistent Patterns (SCP) for cross-view data, which embeds the complementary information from different views, is of great importance for those applications. However, the heterogeneity among cross-view representations brings a significant challenge on mining the SCP. In this paper, we propose a general framework to discover the SCP for cross-view data. Specifically, aiming at building a feature-isomorphic space among different views, a novel Isomorphic Relevant Redundant Transformation (IRRT) is first proposed. The IRRT linearly maps multiple heterogeneous low-level feature spaces to a high-dimensional redundant feature-isomorphic one, which we name as mid-level space. Thus, much more complementary information from different views can be captured. Furthermore, to mine the semantic consistency among the isomorphic representations in the mid-level space, we propose a new Correlation-based Joint Feature Learning (CJFL) model to extract a unique high-level semantic subspace shared across the feature-isomorphic data. Consequently, the SCP for cross-view data can be obtained. Comprehensive experiments on three data sets demonstrate the advantages of our framework in classification and retrieval.

Automatic TV commercial categorization is the key component of a smart commercial content analysis and understanding system. In this paper, we focus our research on commercial categorization by mining the temporal-spatial context of TV commercial. To boost the discrimination ability of the traditional VBoW in commercial categorization, a more flexible representation, i.e. sparse coding based VBoW (sVBoW), is presented to describe the co-occurrence of semantic units in different kinds of commercials. Meanwhile, both the temporal and spatial cues are fully exploited to reflect the context characteristics of TV commercial. The promising experimental results show the effectiveness of the proposed scheme on commercial categorization.

The neighborhood structure can significantly impact the effectiveness of image search, and fulfilling the reversibility of neighborhood may improve the image search quality. This paper proposes an effective and efficient scheme for reconstructing the symmetry relationship of k-nearest neighborhood (KNN). In particular, we design a verifying function to learn the prior knowledge of neighborhood reversibility among images. By exploiting the prior knowledge, the image search system will give higher rank to those images that satisfy the reversibility of KNN relationship with the query. In addition, we systematically investigate the sensitivity of neighborhood size on image search quality and propose an adaptive selection scheme for improving robustness of neighborhood reversibility learning methods. The extensive experimental results show that the proposed scheme remarkably improves the image search quality and give a comparable but more stable performance to the state-of-the-art method for various image datasets.

As one of the most successful approaches for recommendation, matrix factorization based Collaborative Filtering (CF) technique has received considerable attentions over the past years. In this paper, we propose an orthogonal matrix factorization model with graph regularization to preserve the consistency of the local structure both in user and item spaces, respectively. Instead of traditional alternating optimization method, a greedy sequential one is introduced to optimize a pair of coupled factor vector and its corresponding loading vector simultaneously each time, thus the original optimization problem is converted into the well-studied Multivariate Eigen Problem (MEP). Furthermore, multiple pairs of coupled eigen-vectors can be obtained in sequence. To guarantee nonrecurring of repetition of solutions, a novel dual-deflation technique is developed to incorporate into the sequential optimization. Experimental results on MovieLens and Each Movie data sets demonstrate that the proposed method is much more competitive compared with the state of the art matrix factorization based collaborative filtering methods.

Proximal support vector machine (PSVM) is a simple but effective classifier, especially for solving large-scale data classification problems. An inherent deficiency of PSVM lies on its inefficiency for dealing with high-dimensional data. In this paper, we propose a parallel version of PSVM (PPSVM). Based on random dimensionality partitioning, PPSVM can obtain partitioned local model parameters in parallel, with combined parameters to form the final global solution. In fact, PPSVM enjoys two properties: 1) It can calculate model parameters in parallel and is therefore a fast learning method with theoretically proved convergence; and 2) It can avoid the inversion of large matrix, which makes it suitable for high-dimensional data. In the paper, we also propose a random PPSVM with randomly partitioned data in each iteration to improve the performance of PSVM. Experimental results on real-world data demonstrate that the proposed methods can obtain similar or even better prediction accuracy than PSVM with much better runtime efficiency.

Independent Component Analysis (ICA) is an effective unsupervised tool to learn statistically independent representations. However, ICA is not only sensitive to whitening but also difficult to learn an over-complete basis set. Consequently, ICA with soft Reconstruction cost(RICA) was presented to learn sparse representations with over-complete basis even on unwhitened data. Nevertheless, this model may not be an optimal discriminative model for classification tasks, because it failed to consider the association between the training sample and its class. In this paper, we propose a supervised Discriminative ICA model with Reconstruction constraint for image classification, named DRICA. DRICA brings in class information to learn the over-complete basis by incorporating inhomogeneous representation cost constraint into the RICA framework. This constraint leads to partition the set of basis vectors into several subsets corresponding to the sample classes, where each subset could sparsely model data samples from the same class but not others. Therefore, the proposed ICA model can learn an over-complete basis and an optimal multi-class classifier jointly. Some experiments carried out on several standard image databases validate the effectiveness of DRICA for image classification.

Active learning traditionally assumes that labeled and unlabeled samples are subject to the same distributions and the goal of an active learner is to label the most informative unlabeled samples. In reality, situations may exist that we may not have unlabeled samples from the same domain as the labeled samples (i.e. target domain), whereas samples from auxiliary domains might be available. Under such situations, an interesting question is whether an active learner can actively label samples from auxiliary domains to benefit the target domain. In this paper, we propose a transfer active learning method, namely Transfer Active SVM (TrAcSVM), which uses a limited number of target instances to iteratively discover and label informative auxiliary instances. TrAcSVM employs an extended sigmoid function as instance weight updating approach to adjust the models for prediction of (newly arrived) target data. Experimental results on real-world data sets demonstrate that TrAcSVM obtains better efficiency and prediction accuracy than its peers.

Content-based video copy detection is very important for copyright protection in view of the growing popularity of video sharing websites, which deals with not only whether a copy occurs in a query video stream but also where the copy is located and where the copy is originated from. While a lot of work has addressed the problem with good performance, less effort has been made to consider the copy detection problem in the case of a continuous query stream, for which precise temporal localization and some complex video transformations like frame insertion and video editing need to be handled. We attempt to attack the problem by presenting a frame fusion based copy detection approach, which converts video copy detection to frame similarity search and frame fusion under a temporal consistency assumption. Our work focuses mainly on the frame fusion stage due to its critical role in copy detection performance. The proposed frame fusion scheme is based on a Viterbi-like algorithm, comprising an online back-tracking strategy with three relaxed constraints. The experimental results show that the proposed approach achieves high localization accuracy in both the query stream and the reference database even when a query video stream undergoes some complex transformations, while achieving comparable performance compared with state-of-the-art copy detection methods.

Automatic TV commercial block detection (CBD) and commercial block segmentation (CBS) are two key components of a smart commercial digesting system. In this paper, we focus our research on CBD and CBS by the means of collaborative exploitation of visual-audio-textual characteristics embedded in commercials. Rather than utilizing exclusively visual-audio characteristics like most previous works, an abundance of textual characteristics associated with commercials are fully exploited. Additionally, Tri-AdaBoost, an interactive ensemble learning manner, is proposed to form a consolidated semantic fusion across visual, audio, and textual characteristics. In order to segment a detected commercial block into multiple individual commercials, additional informative descriptors including textual characteristics are introduced to boost the robustness in the detection of frame marked with product information (FMPI). Together with the characteristics of audio spectral variation pointer and silent position, FMPI can provide a kind of complementary representation architecture to model the similarity of intra-commercial and the dissimilarity of inter-commercial. Experiments are conducted on a large video dataset from both China central television (CCTV) channels and TRECVID'05, and promising experimental results show the effectiveness of the proposed scheme.

Traditional machine learning methods, such as Support Vector Machines (SVMs), usually assume that training and test data share the same distributions. Due to the inherent dynamic data nature, it is often observed that (1) the volumes of the training data may gradually grow; and (2) the existing and the newly arrived samples may be subject to different distributions or learning tasks. In this paper, we propose a Transfer Incremental Support Vector Machine(TrISVM), with the objective of tackling changes in data volumes and learning tasks at the same time. By using new updating rules to calculate the inverse matrix, TrISVM solves the existing incremental learning problem more efficiently, especially for high dimensional data. Furthermore, when using new samples to update the existing models, TrISVM employs sample-based weight adjustment procedures to ensure that the concept transferring between auxiliary and target samples can be leveraged to fulfill the transfer learning goal. Experimental results on real-world data sets demonstrate that TrISVM achieves better efficiency and prediction accuracy than both incremental-learning and transfer-learning based methods. In addition, the results also show that TrISVM is able to achieve bidirectional knowledge transfer between two similar tasks.

Automatic TV commercial detection has become an indispensable part of content-based video analysis technique due to the explosive growth in TV commercial volume. In this paper, a multi-modal (i.e. visual, audio and textual modalities) commercial digesting scheme is proposed to alleviate two challenges in commercial detection, which are the generation of mid-level semantic descriptor and the application of effective discrimination method. Compared with the general program, some unique semantic characteristics are purposely embedded in the commercial to grasp more attention from audience. Aiming at exploring the power of these semantic characteristics, a kind of novel commercial-oriented descriptor from textual modality is proposed, besides taking advantage of those commonly used description means in light of audio and visual modalities. To boost the ability of discrimination of commercial from general program in multi-modal representation space, Tri-AdaBoost, a self-learning method by an interactive way across multiple modalities, is introduced to form a final consolidated decision for discrimination. Moreover, a heuristic post processing strategy based on the temporal consistency is taken to further reduce the false alarms. The promising experimental results show the effectiveness of the proposed scheme with respect to large video data collections.

Analysis on click-through data from a very large search engine log shows that users are usually interested in the top-ranked portion of returned search results. Therefore, it is crucial for search engines to achieve high accuracy on the top-ranked documents. While many methods exist for boosting video search performance, they either pay less attention to the above factor or encounter difficulties in practical applications. In this paper, we present a flexible and effective reranking method, called CR-Reranking, to improve the retrieval effectiveness. To offer high accuracy on the top-ranked results, CR-Reranking employs a cross-reference (CR) strategy to fuse multimodal cues. Specifically, multimodal features are first utilized separately to rerank the initial returned results at the cluster level, and then all the ranked clusters from different modalities are cooperatively used to infer the shots with high relevance. Experimental results show that the search quality, especially on the top-ranked results, is improved significantly.

Generally speaking, several aspects related to relevance feedback based CBIR include what means should be adopted for approximate semantic description of image content, what strategies be applied to sample labeling in feedback and what relevance model would be built for online discrimination. Using random sampling strategy, we construct a set of random subspaces for learning multiple intrinsic descriptions of image content, with each of which stable component classifier can be trained. To enhance the generalization capability of relevance model, the diversified active learning is carried out by collecting more informative samples, i.e. those samples spreading around decision boundary dispersedly. The final favorable performance also contributes to the application of ensemble scheme on individual component classifier.

Automatic image annotation (AIA) is a promising way to improve the performance of image retrieval. In this paper, we propose a novel AIA scheme based on multiple-instance learning (MIL). By introducing the minimum reference set (MRS) into MIL (denoted by MRS-MIL), the positive instances (i.e. regions in images) embedded in the positive bags (i.e. images) can be picked out via reliable inferring for a concept. Generated through the 1-NN classifier, MRS denotes the set of minimum number of bags that correctly classify all the labeled bags. Following the principle of structure risk minimum, MRS shows good generalization ability and is particularly suitable for the problem of being short of labeled training bags, i.e. problem of small samples. Compared with the previous annotation approaches, the experimental results demonstrate that the proposed MRS-MIL based annotation scheme achieves better performance of AIA even with a small set of labeled bags.

Based on multi-cue integration and hierarchical SVM, we present a sequential architecture for efficient car detection under complex outdoor scene in this paper. On the low level, two novel area templates based on edge and interest-point cues respectively are first constructed, which can be applied to forming the identities of visual perception to some extent and thus utilized to reject rapidly most of the negative non-car objects at the cost of missing few of the true ones. Moreover on the high level, both global structure and local texture cues are exploited to characterize the car objects precisely. To improve the computational efficiency of general SVM, a solution approximating based two-level hierarchical SVM is proposed. The experimental results show that the integration of global structure and local texture properties provides more powerful ability in discrimination of car objects from non-car ones. The final high detection performance also contributes to the utilizing of two novel low level visual cues and the hierarchical SVM.

Automatic detection of commercials embedded in digital video materials is a challenging task. In this paper, a three-phase system of commercial detection is constructed. Firstly, video shots and audio cuts are detected respectively and scene boundaries are refined by fusing the audio and video information simultaneously. Then, a SVM classifier is applied to classifying these shots as program shots or commercial ones. Finally, in order to make the commercial segments more accurate, the identity of time-spatial content of commercial shots is fully exploited to eliminate those falsely classified commercial ones, based on which the individual commercial shots can be integrated into the commercial sequences ultimately. The final test results on 30 clips of TV videos verify the practicality of the constructed system.

In this paper, a novel two-stage object recognition algorithm is proposed. As an iterative line search optimization method, the mean shift technique is used for fast generalizing of a set of hypothesis. During the hypothesis generalization procedure, the weighted global shape context is integrated with weighted gray histogram to enhance object representation. As a measure for the discriminative power of probability distributions, the symmetric discrete KL divergence is adopted instead of Bhattacharyya coefficient. In order to handle the problem of negative weights for samples, a new weight regulation method is introduced. For the verification stage, a robust circular Gabor-based object matching algorithm using weighted Hausdorff distance is adopted to give final verification for the set of hypothesis.