Search

The Social Internet of Things (SIoT) is presented as a new paradigm of the Internet of Things that solves the problems of network navigability and provides enhanced service discovery and composition. It aims to socialize the IoT devices and allow them to interact just like humans by creating multiple social relationships. In SIoT scenarios, a device can offer multiple services, and different devices can offer the same services with different parameters and factors of interest, which leads to data sparsity and sheer volume of services. However, this sheer volume of available services makes it difficult for devices to navigate and select the ones that best fit their needs or preferences. On the other hand, the heterogeneous nature and dynamic connectivity of SIoT networks raise the cold start problem in service recommendations. Few works explored the integration of trust-aware approaches with latent feature mining in the SIoT recommendation systems. To address these challenges, we proposed a hybrid latent feature mining and trust-aware model to provide a tailored service recommendation in the SIoT environment. Experimental results conducted on a public dataset reveal the increase of service recommendation accuracy and highlight the proposed framework’s effectiveness in meeting recommendation needs within the scope of SIoT environment.

Most cross-view image matching algorithms focus on designing network structures with excellent performance, ignoring the content information of the image. At the same time, there are non-fixed targets such as cars, ships, and pedestrians in ground perspective images and aerial perspective images. Differences in perspective, direction, and scale cause serious interference with the cross-view matching process. This paper proposes a cross-view image matching method with feature enhancement, which first transforms the empty image to generate a transformation image aligned with the ground–aerial image domain to establish a preliminary geometric correspondence between the ground-space image. Then, the rich feature information of the deep network and the edge information of the cross-convolution layer are used to establish the feature correspondence between the ground-space images. The feature fusion module enhances the tolerance of the network model to scale differences, improving the interference problem of transient non-fixed targets on the matching performance in the images. Finally, the maximum pooling and feature aggregation strategies are adopted to aggregate local features with obvious distinguishability into global features to complete the accurate matching between ground images. The experimental results show that the proposed method has good advance and high accuracy on CVUSA, which is commonly used in public datasets, reaching 92.23%, 98.47%, and 99.74% on the top 1, top 5 and top 10 indicators, respectively, outperforming the original method in the dataset with a limited field of view and image center, better completing the cross-perspective image matching task. [ABSTRACT FROM AUTHOR]

Most cross-view image matching algorithms focus on designing network structures with excellent performance, ignoring the content information of the image. At the same time, there are non-fixed targets such as cars, ships, and pedestrians in ground perspective images and aerial perspective images. Differences in perspective, direction, and scale cause serious interference with the cross-view matching process. This paper proposes a cross-view image matching method with feature enhancement, which first transforms the empty image to generate a transformation image aligned with the ground–aerial image domain to establish a preliminary geometric correspondence between the ground-space image. Then, the rich feature information of the deep network and the edge information of the cross-convolution layer are used to establish the feature correspondence between the ground-space images. The feature fusion module enhances the tolerance of the network model to scale differences, improving the interference problem of transient non-fixed targets on the matching performance in the images. Finally, the maximum pooling and feature aggregation strategies are adopted to aggregate local features with obvious distinguishability into global features to complete the accurate matching between ground images. The experimental results show that the proposed method has good advance and high accuracy on CVUSA, which is commonly used in public datasets, reaching 92.23%, 98.47%, and 99.74% on the top 1, top 5 and top 10 indicators, respectively, outperforming the original method in the dataset with a limited field of view and image center, better completing the cross-perspective image matching task.

Sharp edges and corners are crucial features for high-quality and fine-detailed 3D meshes, which we tend to treat as noises in previous tasks of mesh denoising mistakenly. A challenge arises on how to handle both surfaces and features simultaneously in 3D mesh denoising. Classical works mainly focused on surfaces, whereas features also reasonably need proper processes. In this paper, we propose a feature-aware trilateral filter under the framework of energy minimization for 3D mesh denoising and address the above challenge. Concretely, we treat the challenge as an energy minimization model, where the data and smooth terms are both carefully designed. Apart from this, we introduce a feature-aware trilateral filter for high-quality mesh guidance to the model. In this filter, features are detected and distinguished from surfaces for consequent guidance. With the help of the model and filter, more priors are involved, and we can make global optimization for better denoising performance. We perform experiments on both synthetic and scanned meshes, where both subjective and objective evaluations are displayed to show the superior performance of our method to state-of-the-art methods. Furthermore, two experiments, including ablation tests and parameter sensitivities tests, are conducted to show the robustness and efficiency of our method thoroughly. All these results demonstrate that our method is suitable for robust and efficient performance in future mesh-oriented applications.

Interactive segmentation is a technique for picking objects of interest in images according to users’ input interactions. Some recent works take the users’ interactive input to guide the deep neural network training, where the users’ click information is utilized as weak-supervised information. However, limited by the learning capability of the model, this structure does not accurately represent the user’s interaction intention. In this work, we propose a multi-click interactive segmentation solution for employing human intention to refine the segmentation results. We propose a coarse segmentation network to extract semantic information and generate rough results. Then, we designed a feature-aware attention module according to the symmetry of user intention and image semantic information. Finally, we establish a refinement module to combine the feature-aware results with coarse masks to generate precise intentional segmentation. Furthermore, the feature-aware module is trained as a plug-and-play tool, which can be embedded into most deep image segmentation models for exploiting users’ click information in the training process. We conduct experiments on five common datasets (SBD, GrabCut, DAVIS, Berkeley, MS COCO) and the results prove our attention module can improve the performance of image segmentation networks.

• A novel feature-aware network (FaNet) is proposed, which is applied to few-shot classification for strip steel surface defects. • Two feature-aware modules can effectively extract and enhance the high-dimensional feature information from a few image data. • The first publicly available few-shot classification dataset with large-scale strip steel surface defects is constructed (FSC-20). Accurate classification of surface defects is one of the most important factors in achieving quality inspection for strip steel. Most existing methods are based on fully-supervised learning, which requires a large number of labeled training data samples. In the manufacturing process, collecting defective samples is time-consuming and laborious. So it is very difficult to train a fully supervised model based on few labeled samples. In this paper, we propose a feature-aware network (FaNet) for a few shot defect classification, which can effectively distinguish new classes with a small number of labeled samples. In our proposed FaNet, we use ResNet12 as our baseline. The feature-attention convolution module (FAC) is applied to extract the comprehensive feature information from the base classes, as well as to fuse semantic information by capturing the long-range feature relationships between the upper and lower layers. Meanwhile, during the test phase, an online feature-enhance integration module (FEI) is adopted to average the noise from the support set and query set defect images, further enhancing image features among the different tasks. In addition, we construct a large-scale strip steel surface defects few shot classification dataset (FSC-20) with 20 different types. Experimental results show that the proposed method achieves the best performance compared to state-of-the-art methods for the 5-way 1-shot and 5-way 5-shot tasks. The dataset and code are available at: https://github.com/VDT-2048/FSC-20. [ABSTRACT FROM AUTHOR]

In the area of 3D digital engineering and 3D digital geometry processing, shape simplification is an important task to reduce their requirement of large memory and high time complexity. By incorporating the content-aware visual salience measure of a polygonal mesh into simplification operation, a novel feature-aware shape simplification approach is presented in this paper. Owing to the robust extraction of relief heights on 3D highly detailed meshes, our visual salience measure is defined by a center-surround operator on Gaussian-weighted relief heights in a scale-dependent manner. Guided by our visual salience map, the feature-aware shape simplification algorithm can be performed by weighting the high-dimensional feature space quadric error metric of vertex pair contractions with the weight map derived from our visual salience map. The weighted quadric error metric is calculated in a six-dimensional feature space by combining the position and normal information of mesh vertices. Experimental results demonstrate that our visual salience guided shape simplification scheme can adaptively and effectively re-sample the underlying models in a feature-aware manner, which can account for the visually salient features of the complex shapes and thus yield better visual fidelity. [ABSTRACT FROM AUTHOR]

Volumetric PolyCube-Map-based methods offer automatic ways to construct all-hexahedral meshes for closed 3D polyhedral domains, but their meshing quality is limited by the lack of interior singularities and feature alignment. In the presented work, we propose cut-enhanced PolyCube-Maps, to introduce essential interior singularities and preserve most input features. Our main idea is simple and intuitive: by inserting proper parameterization seams into the initial PolyCube-Map via novel PolyCube cutting operations, the mapping distortion can be reduced significantly. The cut-enhanced PolyCube-Map computation includes feature-aware PolyCube-Map construction and cut-enhanced PolyCube deformation. The former aims to preserve input feature edges during the initial PolyCube-Map construction. The latter introduces seams into the volumetric PolyCube shape by cutting it through selective PolyCube edges and deforms the modified PolyCube under the seamless constraints to compute a low-distortion PolyCube-Map. The hexahedral mesh induced by the final PolyCube-Map can be further enhanced by our mesh improvement algorithm. We validate the efficacy of our method on a collection of more than one hundred CAD models and demonstrate its advantages over other automatic all-hex meshing methods and padding strategies. The limitations of cut-enhanced PolyCube-Maps are also discussed thoroughly. [ABSTRACT FROM AUTHOR]