Search

Image super-resolution (SR) reconstruction is essentially an ill-posed problem, so it is important to design an effective prior. For this purpose, we propose a novel image SR method by learning both non-local and local regularization priors from a given low-resolution image. The non-local prior takes advantage of the redundancy of similar patches in natural images, while the local prior assumes that a target pixel can be estimated by a weighted average of its neighbors. Based on the above considerations, we utilize the non-local means filter to learn a non-local prior and the steering kernel regression to learn a local prior. By assembling the two complementary regularization terms, we propose a maximum a posteriori probability framework for SR recovery. Thorough experimental results suggest that the proposed SR method can reconstruct higher quality results both quantitatively and perceptually. [ABSTRACT FROM AUTHOR]

Until now, neighbor-embedding-based (NE) algorithms for super-resolution (SR) have carried out two independent processes to synthesize high-resolution (HR) image patches. In the first process, neighbor search is performed using the Euclidean distance metric, and in the second process, the optimal weights are determined by solving a constrained least squares problem. However, the separate processes are not optimal. In this paper, we propose a sparse neighbor selection scheme for SR reconstruction. We first predetermine a larger number of neighbors as potential candidates and develop an extended Robust-SL0 algorithm to simultaneously find the neighbors and to solve the reconstruction weights. Recognizing that the k-nearest neighbor (k-NN) for reconstruction should have similar local geometric structures based on clustering, we employ a local statistical feature, namely histograms of oriented gradients (HoG) of low-resolution (LR) image patches, to perform such clustering. By conveying local structural information of HoG in the synthesis stage, the k-NN of each LR input patch is adaptively chosen from their associated subset, which significantly improves the speed of synthesizing the HR image while preserving the quality of reconstruction. Experimental results suggest that the proposed method can achieve competitive SR quality compared with other state-of-the-art baselines. [ABSTRACT FROM PUBLISHER]

The neighbor-embedding (NE) algorithm for single-image super-resolution (SR) reconstruction assumes that the feature spaces of low-resolution (LR) and high-resolution (HR) patches are locally isometric. However, this is not true for SR because of one-to-many mappings between LR and HR patches. To overcome or at least to reduce the problem for NE-based SR reconstruction, we apply a joint learning technique to train two projection matrices simultaneously and to map the original LR and HR feature spaces onto a unified feature subspace. Subsequently, the k-nearest neighbor selection of the input LR image patches is conducted in the unified feature subspace to estimate the reconstruction weights. To handle a large number of samples, joint learning locally exploits a coupled constraint by linking the LR–HR counterparts together with the K-nearest grouping patch pairs. In order to refine further the initial SR estimate, we impose a global reconstruction constraint on the SR outcome based on the maximum a posteriori framework. Preliminary experiments suggest that the proposed algorithm outperforms NE-related baselines. [ABSTRACT FROM AUTHOR]

Transfer learning is an effective means by leveraging rich information of the labeled samples from the source domain to build an appropriate classifier for annotating the target domain. Existing transfer learning algorithms usually try to minimize the distribution difference between different domains by means of Maximum Mean Discrepancy (MMD) based metric criteria. However, the domain adaptation methods based on MMD usually have the following two problems: 1) the local geometry information of the inter-domain data is ignored, which may cause to negative transfer effect. 2) The classifier and the feature learning are not integrated into one framework, in which the global optimal solution may be not obtained. Therefore, a Robust Manifold Discriminative Distribution Adaptation (RMDDA) model is presented for transfer subspace learning. Because the difference between domains leads to the large deviation of the original features of data in the source and target domains, it is difficult to construct the manifold structure of inter-domain data. A new feature termed Histogram Feature of Neighbors (HFON) is firstly introduced to rightly capture the domain locality. Secondly, the marginal distribution, conditional distribution, and discriminative regression are integrated into a model, which can solve the problem of separating feature learning from classifier training. It is considered that the dimension of feature extraction is limited by the number of classes which is determined by discriminative regression, we introduce double projection matrices into the model. Lastly, the numerous experiments on six benchmark cross-domain image datasets prove the superiority of RMDDA algorithm compared with some state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

In recent years, the Transformer model has made significant progress in semantic segmentation tasks. However, existing self-attention mechanisms perform well in capturing remote dependencies and global features, but ignore local area information in point cloud data and have limitations in dealing with multi-scale features. To address this problem, this paper introduces a multiscale self-attention fusion (MSA) module, which adaptively fuses features within different scale neighborhoods and learns global contextual features by connecting local neighborhoods. Then, the multiscale channel aggregation module (MCA)is used to perform deep point-by-point and point-by-point convolution of the point cloud channel, aggregating channel features at multiple scales to extract more accurate local feature information. Finally, in this study, the multiscale adaptive fusion (MSA) module and the multiscale channel aggregation (MCA) module form a sequential network structure that adaptively and dynamically adjusts different scales of point cloud objects to enhance the perception of objects of different sizes for better segmentation performance. By testing and validating the model on the publicly available S3DIS Area 5 dataset and the ScantNetV2 dataset, the model achieves mIoU index values of 71.9% and 72.3%, respectively, which demonstrates the effectiveness and superiority of the proposed method. Code will be made publicly available at https://github.com/Cocoyufei/MAT/tree/master. [Display omitted] [ABSTRACT FROM AUTHOR]

Low-light image enhancement strives to improve the contrast, adjust the visibility, and restore the distortion in color and texture. Existing methods usually pay more attention to improving the visibility and contrast via increasing the lightness of low-light images, while disregarding the significance of color and texture restoration for high-quality images. Against above issue, we propose a novel luminance and chrominance dual branch network, termed LCDBNet, for low-light image enhancement, which divides low-light image enhancement into two sub-tasks, e.g., luminance adjustment and chrominance restoration. Specifically, LCDBNet is composed of two branches, namely luminance adjustment network (LAN) and chrominance restoration network (CRN). In LAN, we design a global and local aggregation block (GLAB) to extract brightness-aware features, which consists of a transformer branch and a dual attention branch to model long-range dependency and local attention correlation. In CRN, we introduce wavelet transform to obtain high-frequency detail information. Finally, a fusion network is designed to blend their learned features to produce visually impressive images. Extensive experiments conducted on seven benchmark datasets validate the effectiveness of our proposed LCDBNet, and the results manifest that LCDBNet achieves superior performance in terms of multiple reference/non-reference quality evaluators compared to other state-of-the-art competitors. Our code and pretrained model are available at https://github.com/WHK-Huake/LCDBNet. [ABSTRACT FROM AUTHOR]

To synthesize the available evidence on factors associated with self-management behavior in young stroke patients. The methodological guidelines for scoping reviews developed by the Joanna Briggs Institute and the PRISMA-scR-checklist for scoping reviews were used. A total of 5586 studies were identified through bibliographic searches of the scientific databases Medline (OVID), Embase (OVID), CINAHL (EBSCO), and PsycINFO, limited to the period 2000–2023. Studies were independently assessed for inclusion and exclusion criteria by two reviewers. Quantitative observational data and qualitative studies were extracted, mapped, and summarized to provide a descriptive summary of trends and considerations for future research. Nine papers were finally selected to answer the research question. Young patients' self-management was mainly influenced by demographic factors (age, gender, income, education, and stroke knowledge), disease-related factors (functionality and independence, duration of stroke diagnosis, cognitive function, and poststroke fatigue), and psychosocial factors (hardiness, spiritual self-care, self-efficacy, and social support). Further research is needed to determine the trajectory of poststroke self-management over time and its potential predictors, which should lead to the development of specific stroke rehabilitation and stroke self-management support programs for young people (considering factors that influence return to work in young stroke patients' self-management). Healthcare providers can design more efficient interventions to improve the quality of life of young stroke patients after discharge. Gaining an in-depth understanding of the factors that influence self-management can help achieve this. ● The incidence of stroke among employed individuals aged 18-65 has increased by 44% over the past decade. ● Lack of research on factors affecting self-management of stroke survivors in young and middle-aged. ● Self-management in young and middle-aged stroke survivors is influenced by demographic, disease-related, and psychosocial factors.: ● The factors can provide valuable information for treatment decisions, prognosis, and the design of patient-centered services. [ABSTRACT FROM AUTHOR]

In the paper, we present a new transfer subspace learning algorithm termed discriminative transfer regression (DTR) for cross-domain image recognition, in which low-rank representation (LRR), discriminative regression, local geometry preserving, and different norm minimization are integrated in to a united framework for transfer learning. Firstly, both the global structure and the local geometry information of the original data are preserved by imposing low-rank and sparse constraints on the reconstruction coefficient matrix. Secondly, DTR algorithm can overcome the disturbance of outliers and noises. Thirdly, the local manifold structure of the observed samples with the same semantics from the source and target domains is captured by the adaptive weight graph. Fourthly, the discriminative information of the samples from the source domain is encoded to the target domain based on ridge regression (RR). Meanwhile, the small-class problem confronted by RR and its extensions, that the obtained projection matrix is limited by the number of classes, can be effectively solved. In addition, the convergence and computational complexity of DTR algorithm is analyzed. An extensive range of experiments on several cross-domain image databases demonstrate the superiority of the DTR algorithm. [ABSTRACT FROM AUTHOR]

In the context of cross-modal person re-identification, researchers often employ methods that utilize visible modality information to generate both an 'X' modality and a grayscale modality, enhancing the accuracy of person re-identification models. A lightweight network causes the 'X' modality through self-supervised learning of labels from visible images. In contrast, the grayscale modality is obtained through simple linear accumulation of the three RGB color channels from visual images. It can be observed that both the 'X' modality and grayscale modality are derived from visible images, which fails to establish a connection between the visible and infrared modalities. Therefore, this paper proposes an intermediate modality generation module to produce intermediate modality representations dynamically. By combining information from the visible, infrared, and intermediate modalities, the model is encouraged to capture modality-invariant features with cross-modal consistency. This enables person of the same identity to exhibit similar feature representations across different modalities, thereby mitigating the impact of distribution differences between visible and infrared modalities. Additionally, to facilitate the learning of appropriate intermediate modality representations, a distribution migration strategy is introduced. It guides the intermediate modality to maintain the correct distance from the visible and infrared modalities by optimizing the weights of the loss functions, preventing inadequate feature learning caused by an excessive focus on specific modality information. Furthermore, a mixed augmentation approach is proposed to alleviate disparities among multiple modalities further. By randomly cropping and combining regions of visible (infrared) modality images with infrared (visible) modality images, the generalization ability of the model in heterogeneous modalities is enhanced. Extensive comparative experiments are conducted on the SYSU-MM01 and RegDB datasets, yielding mAP values of 57.2% and 85.82%, respectively. The superior mAP performance on the RegDB dataset compared to most existing methods validates the effectiveness of the proposed approach. • Significant differences in data distribution between different modalities • X modality generated solely from visible images cannot be linked to infrared modality • Joint learning of visible modality, infrared modality, and intermediate modality • The distributed migration strategy can learn richer modal features [ABSTRACT FROM AUTHOR]

For point cloud semantic segmentation, it makes sense to focus on the boundary information of transition regions. However, most existing excellent point cloud segmentation networks tend to overlook boundary information, lead to misclassification in transition areas. For this reason, we propose a network called Boundary-aware dual edge network (BADENet). Firstly, the backbone network is optimized to make the extracted features more robust. Secondly, a boundary-aware module is utilized to focus on the target boundary in the transition region. This module contains a boundary point prediction module and a feature aggregation module. The boundary point prediction module predicts the points belonging to the target boundary by learning the neighborhood point features. The feature aggregation module performs discriminative aggregation of point cloud features in the local neighborhood under the guidance of the target boundary. In addition, an effective dual edge convolution module is designed to model the graph topology, which captures fine-grained geometric information through edge convolution operations, further improves the semantic feature recognition capability. Finally, the effectiveness of the method is verified on the public datasets ScanNetV2 and S3DIS. The mIoU reaching 64.9% on ScanNetV2, and the OA reaching 88.7% on S3DIS. Code will be made publicly available at https://github.com/jietry/zm [ABSTRACT FROM AUTHOR]

Image-based virtual try-on involves generating an image of a person wearing a given clothing. Existing virtual try-on works suffer from the problem of misaligned regions between the predicted segmentation map and the deformed clothing, and the generation results of try-on are unnatural. To address this issue, we refine the definition of the misaligned regions and propose a high-resolution virtual try-on network called CS-VITON. The network adopts a two-stage strategy. The first stage is called the condition generator, which predicts the target segmentation map while deforming the clothing into shapes that match the human body. A component that measures the difference between the generated segmentation maps and the mask of deformed clothing is added to the loss function of the deep network. The component is well matched with the tasks of this stage, resulting in more reasonable necklines and skin boundaries. The second stage is called the try-on generator, in which the process of generating try-on images is modulated using residual blocks constructed based on style-preserved modulation. The modulation process takes into account the specific contextual style of the image, which improves the realism of the try-on results. Extensive experiments were conducted on a common high-resolution virtual try-on dataset, demonstrating that that our method yields more realistic virtual try-on results. Metrics such as kernel inception distance also showed some improvement. The code will be available soon at https://github.com/xinz626/CS-VITON. [ABSTRACT FROM AUTHOR]

Example learning-based superresolution (SR) algorithms show promise for restoring a high-resolution (HR) image from a single low-resolution (LR) input. The most popular approaches, however, are either time- or space-intensive, which limits their practical applications in many resource-limited settings. In this paper, we propose a novel computationally efficient single image SR method that learns multiple linear mappings (MLM) to directly transform LR feature subspaces into HR subspaces. In particular, we first partition the large nonlinear feature space of LR images into a cluster of linear subspaces. Multiple LR subdictionaries are then learned, followed by inferring the corresponding HR subdictionaries based on the assumption that the LR-HR features share the same representation coefficients. We establish MLM from the input LR features to the desired HR outputs in order to achieve fast yet stable SR recovery. Furthermore, in order to suppress displeasing artifacts generated by the MLM-based method, we apply a fast nonlocal means algorithm to construct a simple yet effective similarity-based regularization term for SR enhancement. Experimental results indicate that our approach is both quantitatively and qualitatively superior to other application-oriented SR methods, while maintaining relatively low time and space complexity.