Search

Sequential patterns are important, because they can be exploited to improve the prediction accuracy of our classifiers. Sequential data, such as time series/video frames, and event data are becoming more and more ubiquitous in a wide spectrum of application scenarios especially in the background of large data and deep learning. However, large data sets used in training modern machine-learning models, such as deep neural networks, are often affected by label noise. Existing noisy learning approaches mainly focus on building an additional network to clean the noise or find a robust loss function. Few works tackle this problem by exploiting sample correlations. In this paper, we propose BundleNet, a framework of sequential structure (named bundle-module, see Fig. 1) for deep neural networks to handle the label noise. The bundle module naturally takes into account sample correlations by constructing bundles of samples class-by-class, and treats them as independent inputs. Moreover, we prove that the bundle-module performs a form of regularization, which is similar to dropout as regularization during training. The regularization effect endows the BundleNet with strong robustness to the label noise. Extensive experiments on public data sets prove that the proposed approach is effective and promising.

Bounding box estimation by overlap maximization has improved the state of the art of visual tracking significantly, yet the improvement in robustness and accuracy is restricted by the limited reference information, i.e., the initial target. In this paper, we present DCOM, a novel bounding box estimation method for visual tracking, based on distribution calibration and overlap maximization. We assume every dimension in the modulation vector follows a Gaussian distribution, so that the mean and the variance can borrow from those of similar targets in large-scale training datasets. As such, sufficient and reliable reference information can be obtained from the calibrated distribution, leading to a more robust and accurate target estimation. Additionally, an updating strategy for the modulation vector is proposed to adapt the variation of the target object. Our method can be built on top of off-the-shelf networks without finetuning and extra parameters. It yields state-of-the-art performance on three popular benchmarks, including GOT-10k, LaSOT, and NfS while running at around 40 FPS, confirming its effectiveness and efficiency.

Efficient indexing and retrieving objects of interest from large‐scale surveillance videos are a significant and challenging topic. In this study, the authors present an effective multiple deep features learning approach for object retrieval in surveillance videos. Based on the discriminative convolutional neural network (CNN), they can learn multiple deep features to comprehensively describe the visual object. To be specific, they utilise the CNN model pre‐trained on ImageNet ILSVRC12 and fine‐tuned on our dataset to abstract structure information. In addition, they train another CNN model supervised by 11 colour names to deliver the colour information. To improve the retrieval performance, the deep features are encoded into short binary codes by locality‐sensitive hash and fused to fast retrieve the object of interest. Retrieval experiments are performed on a dataset of 100k objects extracted from multi‐camera surveillance videos. Comparison results with other common visual features show the effectiveness of the proposed approach.

In this paper, an improvement of deformable template matching algorithm for polygonal objects in grayscale images using two-dimensional deformable templates along orthogonal curves is presented. In the process of pre-computing extensions of the deformable template along orthogonal curves, the novel matching approach incorporates adapting knowledge-specific template discretization techniques appropriate for different polygonal objects and minimizing the improved internal and external energy terms containing inter-shape information of polygonal objects. In our application, this energy optimization problem of the deformable template is efficiently solved by a genetic algorithm (GA). Our algorithm has been successfully applied on synthetic images and real images. The experiment results show that the new approach provides more robust and accurate matching method.

We present an integrated framework on personalized sports video customization, which addresses three research issues: semantic video annotation, personalized video retrieval and summarization, and system adaptation. Sports video annotation serves as the foundation of the video customization system. To acquire detailed description of video content, external web text is adopted to align with the related sports video according to their semantic correspondence. Based on the derived semantic annotation, a user-participant multiconstraint 0/1 Knapsack model is designed to model the personalized video customization, which can unify both video retrieval and summarization with different fusion parameters. As a measure to make the system adaptive to the particular user, a social network based system adaptation algorithm is proposed to learn latent user preference implicitly. Both quantitative and qualitative experiments conducted on twelve broadcast basketball and football videos validate the effectiveness of the proposed method.

In this paper, we present an effective hierarchical shot classification scheme for broadcast soccer video. We first partition a video into replay and non-replay shots with replay logo detection. Then, non-replay shots are further classified into Long, Medium, Close-up or Out-field types with color and texture features based on a decision tree. We tested the method on real broadcast FIFA soccer videos, and the experimental results demonstrate its effectiveness..

Abstract Background Cox‐Maze procedure is currently the gold standard treatment for atrial fibrillation (AF). However, data on the effectiveness of the Cox‐Maze procedure after concomitant mitral valve surgery (MVS) are not well established. The aim of this study was to assess the safety and efficacy of Cox-Maze procedure versus no-maze procedure n in AF patients undergoing mitral valve surgery through a systematic review of the literature and meta‐analysis. Methods A systematic search on PubMed/MEDLINE, EMBASE, and Cochrane Central Register of Clinical Trials (Cochrane Library, Issue 02, 2017) databases were performed using three databases from their inception to March 2023, identifying all relevant randomized controlled trials (RCTs) comparing Cox-Maze procedure versus no procedure in AF patients undergoing mitral valve surgery. Data were extracted and analyzed according to predefined clinical endpoints. Results Nine RCTs meeting the inclusion criteria were included in this systematic review with 663 patients in total (341 concomitant Cox‐Maze with MVS and 322 MVS alone). Across all studies with included AF patients undergoing MV surgery, the concomitant Cox‐Maze procedure was associated with significantly higher sinus rhythm rate at discharge, 6 months, and 12 months follow‐up when compared with the no-Maze group. Results indicated that there was no significant difference between the Cox‐Maze and no-Maze groups in terms of 1 year all-cause mortality, pacemaker implantation, stroke, and thromboembolism. Conclusions Our systematic review suggested that RCTs have demonstrated the addition of the Cox‐Maze procedure for AF leads to a significantly higher rate of sinus rhythm in mitral valve surgical patients, with no increase in the rates of mortality, pacemaker implantation, stroke, and thromboembolism.

IntroductionPeripheral vascular atherosclerosis (PVA) is a chronic inflammatory disease characterized by lipid accumulation in blood vessel walls, leading to vessel narrowing and inadequate blood supply. However, the molecular mechanisms underlying PVA remain poorly understood. In this study, we employed a combination of Mendelian randomization (MR) analysis and integrated transcriptomics to identify the critical gene signature associated with PVA.MethodsThis study utilized three public datasets (GSE43292, GSE100927 and GSE28829) related to peripheral vascular atherosclerosis obtained from the Gene Expression Omnibus database. Instrumental variables (IVs) were identified through expression quantitative trait loci (eQTL) analysis, and two-sample MR analysis was performed using publicly available summary statistics. Disease critical genes were identified based on odds ratios and intersected with differentially expressed genes in the disease dataset. GSE28829 dataset was used to validate the screened disease critical genes. Functional enrichment analysis, GSEA analysis, and immune cell infiltration analysis were performed to further characterize the role of these genes in peripheral vascular atherosclerosis.ResultsA total of 26,152 gene-related SNPs were identified as IVs, and 242 disease-associated genes were identified through MR analysis. Ten disease critical genes (ARHGAP25, HCLS1, HVCN1, RBM47, LILRB1, PLAU, IFI44L, IL1B, IFI6, and CFL2) were significantly associated with peripheral vascular atherosclerosis. Functional enrichment analysis using KEGG pathways revealed enrichment in the NF-kappa B signaling pathway and osteoclast differentiation. Gene set enrichment analysis further demonstrated functional enrichment of these genes in processes related to vascular functions and immune system activation. Additionally, immune cell infiltration analysis showed differential ratios of B cells and mast cells between the disease and control groups. The correlations analysis highlights the intricate interplay between disease critical genes and immune cells associated with PVA.ConclusionIn conclusion, this study provides new insights into the molecular mechanisms underlying PVA by identifying ten disease critical genes associated with the disease. These findings, supported by differential expression, functional enrichment, and immune system involvement, emphasize the role of these genes in vascular function and immune cell interactions in the context of PVA. These findings contribute to a better understanding of PVA pathogenesis and offer potential targets for further mechanistic exploration and therapeutic interventions.

Bilinear models are very powerful in multimodal fusion tasks like Visual Question Answering. The predominant bilinear methods can all be seen as a kind of tensor-based decomposition operation that contains a key kernel called "core tensor." Current approaches usually focus on reducing the computation complexity by applying low-rank constraint on the core tensor. In this article, we propose a novel bilinear architecture called Block Term Decomposition Pooling (BTDP), which not only maintains the advantages of previous bilinear methods but also conducts sparse bilinear interactions between modalities. Our method is based on Block Term Decompositions theory of tensor, which will result in a sparse and learnable block-diagonal core tensor for multimodal fusion. We prove that using such a block-diagonal core tensor is equivalent to conducting many "tiny" bilinear operations in different feature spaces. Thus, introducing sparsity into the bilinear operation can significantly increase the performance of feature fusion and improve VQA models. What is more, our BTDP is very flexible in design. We develop several variants of BTDP and discuss the effects of the diagonal blocks of core tensor. Extensive experiments on two challenging VQA-v1 and VQA-v2 datasets show that our BTDP method outperforms current bilinear models, achieving state-of-the-art performance. [ABSTRACT FROM AUTHOR]

In this paper, we develop a stochastic Asymptotic-Preserving (sAP) scheme for the kinetic chemotaxis system with random inputs, which will converge to the modified Keller--Segel model with random inputs in the diffusive regime. Based on the generalized Polynomial Chaos (gPC) approach, we design a high order stochastic Galerkin method using implicit-explicit (IMEX) Runge--Kutta (RK) time discretization with a macroscopic penalty term. The new schemes improve the parabolic CFL condition to a hyperbolic type when the mean free path is small, which shows significant efficiency especially in uncertainty quantification (UQ) with multiscale problems. The sAP property will be shown asymptotically and verified numerically in several tests. Other numerical tests are conducted to explore the effect of the randomness in the kinetic system, with the goal of providing more intuition for the theoretic study of the chemotaxis models. [ABSTRACT FROM AUTHOR]

For linear transport and radiative heat transfer equations with random inputs, we develop new generalized polynomial chaos based asymptotic-preserving stochastic Galerkin schemes that allow efficient computation for the problems that contain both uncertainties and multiple scales. Compared with previous methods for these problems, our new method uses the implicit-explicit time discretization to gain higher order accuracy, and by using a modified diffusion operator based penalty method, a more relaxed stability condition-a hyperbolic, rather than parabolic, CFL stabil- ity condition-is achieved in the case of a small mean free path in the diffusive regime. The stochastic asymptotic-preserving property of these methods will be shown asymptotically and demonstrated nu- merically, along with a computational cost comparison with previous methods. [ABSTRACT FROM AUTHOR]

A social recommendation system has attracted a lot of attention recently in the research communities of information retrieval, machine learning, and data mining. Traditional social recommendation algorithms are often based on batch machine learning methods which suffer from several critical limitations, e.g., extremely expensive model retraining cost whenever new user ratings arrive, unable to capture the change of user preferences over time. Therefore, it is important to make social recommendation system suitable for real-world online applications where data often arrives sequentially and user preferences may change dynamically and rapidly. In this paper, we present a new framework of online social recommendation from the viewpoint of online graph regularized user preference learning (OGRPL), which incorporates both collaborative user-item relationship as well as item content features into an unified preference learning process. We further develop an efficient iterative procedure, OGRPL-FW which utilizes the Frank-Wolfe algorithm, to solve the proposed online optimization problem. We conduct extensive experiments on several large-scale datasets, in which the encouraging results demonstrate that the proposed algorithms obtain significantly lower errors (in terms of both RMSE and MAE) than the state-of-the-art online recommendation methods when receiving the same amount of training data in the online learning process. [ABSTRACT FROM PUBLISHER]