Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks448 from public data
- 583
- 287
- Augmented Skeleton Based Contrastive Action Learning with Momentum LSTM for Unsupervised Action Recognition236
This paper proposes a contrastive action learning paradigm named AS-CAL that can leverage different augmentations of unlabeled skeleton data to learn action representations in an unsupervised manner and can achieve comparable or even superior performance to numerous supervised learning methods.
- 217
- Data Augmentation and Dense-LSTM for Human Activity Recognition Using WiFi Signal188
A WiFi-based human activity recognition system that synthesizes variant activities data through eight channel state information (CSI) transformation methods to mitigate the impact of activity inconsistency and subject-specific issues is proposed and a novel deep-learning model is designed that caters to the small-size WiFi activity data.
- 166
- Collect and Select: Semantic Alignment Metric Learning for Few-Shot Learning164
A Semantic Alignment Metric Learning (SAML) method for few-shot learning that aligns the semantically relevant dominant objects through a ``collect-and-select'' strategy and demonstrates the generalization ability of SAML and gives a theoretical guarantee.
- 161
- 145
- A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications136
To facilitate sports analytics, a toolbox using PaddlePaddle is developed, which supports football, basketball, table tennis and figure skating action recognition, and discusses the challenges and unsolved problems in this area.
- Pairwise constraints based multiview features fusion for scene classification129
A novel multiview dimensionality reduction method for scene classification that takes both intraclass and interclass geometries into consideration, and has the best performance in scene classification.
- 124
- Sketch guided and progressive growing GAN for realistic and editable ultrasound image synthesis123
The first work that can synthesize realistic B-mode US images with high-resolution and customized texture editing features is presented, and a feature loss is proposed to minimize the difference of high-level features between the generated and real images.
- Canonical correlation analysis networks for two-view image recognition123
A canonical correlation analysis network (CCANet) is presented to address image classification, in which images are represented by two-view features, and the experimental results demonstrate that the CCANet algorithm is more effective than PCANet, RandNet-1 andRandNet-2.
- Handbook on Advancements in Smart Antenna Technologies for Wireless Networks122
This uniquely comprehensive reference offers in-depth descriptions of terminologies, concepts, methods, and applications related to smart antennas in various wireless systems.
- Human Action Recognition by Learning Spatio-Temporal Features With Deep Neural Networks118
This paper proposes one lightweight action recognition architecture based on deep neural networks just using RGB data, consisting of convolution neural network, long short-term memory (LSTM) units, and temporal-wise attention model.
- Semantic preserving distance metric learning and applications116
A Semantic Preserving Distance Metric Learning (SP-DML) algorithm is developed to explore the complementary characteristics of the visual features and pairwise constraints in a unified feature space to integrate semantic contents in distance metric learning.
- 115
- Mobile Cyber Physical Systems: Current Challenges and Future Networking Applications115
This paper distinguishes mobile CPS from traditional CPS, introduces their three emerging application areas, i.e., vehicular networking systems, healthcare systems, and mobile education, and considers the corresponding techniques, which may address these challenges, and analyze the inter-relations among them.
- 108
- 104
- A Self-Supervised Gait Encoding Approach With Locality-Awareness for 3D Skeleton Based Person Re-Identification101
A self-supervised gait encoding approach that can leverage unlabeled skeleton data to learn gait representations for person re-identification via gait features within 3D skeleton sequences and a locality-aware attention mechanism and contrastive learning scheme, which aim to preserve locality-awareness on intra-sequence level and inter- sequence level respectively during self- supervised learning.
- 100
- Multiview dimension reduction via Hessian multiset canonical correlations100
HesMCC is presented, which takes the advantage of Hessian and provides superior extrapolating capability and finally leverage the performance of TCCA, KMUDA, MCCA and LapMCC for multiview dimension reduction.
- Skeleton embedded motion body partition for human action recognition using depth sequences95
An effective approach, which is insensitive to various temporal duration and adequate for complex background, for human action recognition using depth sequences, is proposed by embedding the skeleton information into depth maps by taking account of the geometrical structure of human body and contribute to the recognition task in complex background.
- 93
- Deep features for person re-identification on metric learning92
This work summarizes the different types of features and metric learning approaches from a label attributes perspective, and conducts comprehensive experiments on metric learning methods with two datasets, showing that the relations of loss function with deep feature space and metriclearning.
- 78
- Prototypical Contrast and Reverse Prediction: Unsupervised Skeleton Based Action Recognition76
Extensive experiments on N-UCLA, NTU 60, and NTU 120 dataset present that PCRP outperforms main stream unsupervised methods and even achieves superior performance over many supervised methods.
- 74
- 68
- 66
- 63
- 63
- Depression Prevalence in Postgraduate Students and Its Association With Gait Abnormality61
This paper finds that around 6.5% of the university postgraduate students in China experience depression, and investigates whether the gait patterns of these individuals have already changed as depression is suggested to associate with gait abnormality.
- Multiview Cauchy Estimator Feature Embedding for Depth and Inertial Sensor-Based Human Action Recognition61
A new unsupervised feature fusion method termed multiview Cauchy estimator feature embedding (MCEFE) for human action recognition is proposed, which integrates the encoded complementary information in multiple views to find the unified data representation and the projection matrices.
- Multi-scale spatial–temporal convolutional neural network for skeleton-based action recognition60
A novel multi-scale spatial–temporal convolutional (MSST) module is proposed to take the implicit complementary advantages across spatial-temporal representations with different scales and achieves competitive performance with much less computational cost than state-of-the-art methods.
- Real-Time Semantic Segmentation via Spatial-Detail Guided Context Propagation60
This work proposes the spatial-detail guided context propagation network (SGCPNet), which uses the spatial details of shallow layers to guide the propagation of the low-resolution global contexts, in which the lost spatial information can be effectively reconstructed and improve the model efficiency.
- 58
- The spatial Laplacian and temporal energy pyramid representation for human action recognition using depth sequences56
This work proposes an effective method for human action recognition using depth sequences captured by depth cameras, and demonstrates its advantages over existing methods and is likely to be highly useful for online applications.
- 3D object detection based on sparse convolution neural network and feature fusion for autonomous driving in smart cities54
This paper proposes one end-to-end three dimensional object detection method based on voxelization, sparse convolution, and feature fusion that achieves state-of-the-art performance especially in the Cyclist class.
- 53
- 50
- Diversification of Genes Encoding Granule-Bound Starch Synthase in Monocots and Dicots Is Marked by Multiple Genome-Wide Duplication Events50
Findings suggest that it is GBSSII rather than GBSSI of monocots that have orthologous relationships withGBSS genes of eudicots, and diversification of GBSs genes is mainly associated with genome-wide duplication events throughout the evolutionary course of history ofmonocots and eudicot.
- Multi-modality learning for human action recognition49
An effective segmentation convolutional networks (ConvNets) architecture based on multi-modality hierarchical fusion strategy for human action recognition and a bidirectional rank pooling method to construct the RGB Visual Dynamic Images (VDIs) and Depth Dynamic images (DDIs).
- Feature fusion for 3D hand gesture recognition by learning a shared hidden space48
This paper proposes a novel feature fusion approach for 3D hand gesture recognition that combines the merits of both features, gesture trajectories and acceleration signals, and results confirm the effectiveness of the feature fusion method.
- Reinforcement online learning for emotion prediction by using physiological signals46
A reinforcement online learning (ROL) method for real-time emotion state prediction by exploiting the reward to modify the predictor during the online training iterations is proposed and gains both significant time reduction and prominent performance.
- 45
- 45
- 44
- 43
- 42
- 42
- 41
- Exploiting spatio-temporal representation for 3D human action recognition from depth map sequences41
A straightforward and efficient framework for modeling the human action based on depth map sequences, considering the short-term and long-term dependencies is proposed, and a frame-level feature, termed depth-oriented gradient vector (DOGV), is developed to capture the appearance and motion in a short- term duration.
- 41
- 39
- 39
- Intelligent Smoke Alarm System with Wireless Sensor Network Using ZigBee36
An intelligent smoke alarm system that uses ZigBee transmission technology to build a wireless network, uses random forest to identify smoke, and uses E-charts for data visualization is introduced.
- 36
- Classification-based learning by particle swarm optimization for wall-following robot navigation34
The parameter setting for a set of intelligent multi-category classifiers in wall-following robot navigation is studied and a particle selecting approach is proposed to search for the optimal parameters, a key property of this set of multi- category classifiers.
- 33
- 33
- 33
- 33
- Self-Supervised Gait Encoding with Locality-Aware Attention for Person Re-Identification32
This work proposes a generic gait encoding approach that can utilize unlabeled skeleton data to learn gait representations in a self-supervised manner, and proposes Attention-based Gait Encodings (AGEs), which are built using context vectors learned by locality-aware attention, as final gait representation.
- Segment spatial-temporal representation and cooperative learning of convolution neural networks for multimodal-based action recognition31
The experimental results demonstrate that the proposed segmented rank pooling can represent discriminative spatial-temporal information from the entire RGB and depth sequence, and the proposed SC-ConvNets can enhance recognition performance by learning complementary features from different modalities.
- Robust Sparse Coding for Mobile Image Labeling on the Cloud31
A novel mobile image labeling scheme on the cloud is presented, in which mobile images are first and efficiently transmitted to the cloud by Hamming compressed sensing, such that the heavy computation for image understanding is transferred to thecloud for quick response to the queries of the users.
- Robust leaderless time-varying formation control for unmanned aerial vehicle swarm system with Lipschitz nonlinear dynamics and directed switching topologies30
This paper tackles the robust leaderless Time-Varying Formation (TVF) control problem for the Unmanned Aerial Vehicle (UAV) swarm system with Lipschitz nonlinear dynamics, external disturbances and directed switching topologies with a certain matrix variable substitution.
- LMAE: A large margin Auto-Encoders for classification30
The Large Margin Auto-Encoders (LMAE) is proposed to further boost the discriminability by enforcing different class samples to be large marginally distributed in hidden feature space and demonstrates that the proposed LMAE outperforms the traditional Auto-encoders algorithm.
- 29
- 28
- Replacement of sedentary behavior with various physical activities and the risk of all-cause and cause-specific mortality28
Examining the associations between prolonged sedentary behavior and the risk of all-cause and cause-specific mortality as well as to explore desirable alternatives to sitting in terms of physical activity (PA), demonstrated that substituting sitting with various forms of PA, even for short periods involving relatively light and relaxing physical activity, effectively reduced the risk of both overall and cause-specific mortality.
- Local structure preserving discriminative projections for RGB-D sensor-based scene classification28
A new subspace selection scheme called local structure preserving discriminative projections (LSPDP) is presented, which simultaneously considers two issues: achieving a balance between the local patch structure and global within-class structure and maximizing between-class distances.
- 26
- 26
- Harmonizing Base and Novel Classes: A Class-Contrastive Approach for Generalized Few-Shot Segmentation26
This work proposes a class contrastive loss and a class relationship loss to regulate prototype updates and encourage a large distance between prototypes from different classes, thus distinguishing the classes from each other while maintaining the performance of the base classes.
- Research on task-offloading decision mechanism in mobile edge computing-based Internet of Vehicle26
A task-offloading decision mechanism with particle swarm optimization for MEC-IoV and results show that the proposed offloading strategy can effectively reduce the energy consumption of terminal devices while guarantee the service quality of users.
- 26
- Embedded adaptive cross-modulation neural network for few-shot learning24
An embedded adaptive cross-modulation (EACM) method for few-shot learning which combines the information between support and query examples and improves the generalization ability and robustness of image recognition.
- Color Transfer Pulse-Coupled Neural Networks for Underwater Robotic Visual Systems24
In this scheme, pulse-coupled neural network (PCNN)-based image enhancement and color transfer algorithms are combined to enhance the underwater image to avoid color imbalance and enhance details while suppressing noise.
- 24
- Locally regularized sliced inverse regression based 3D hand gesture recognition on a dance robot24
A new dimension reduction method is developed, locally regularized sliced inverse regression (LR-SIR), to find an effective low dimensional subspace, in which different hand gestures are well separable, following which recognition can be performed by using simple and efficient classifiers.
- SM-SGE: A Self-Supervised Multi-Scale Skeleton Graph Encoding Framework for Person Re-Identification23
A Self-supervised Multi-scale Skeleton Graph Encoding (SM-SGE) framework that comprehensively models human body, component relations, and skeleton dynamics from unlabeled skeleton graphs of various scales to learn an effective skeleton representation for person Re-ID is proposed.
- 3D human posture segmentation by spectral clustering with surface normal constraint23
A new algorithm for partitioning human posture represented by 3D point clouds sampled from the surface of human body formed as a constrained extension of the recently developed segmentation method, spectral clustering (SC).
- 22
- Graph based transductive learning for cartoon correspondence construction22
An automatic cartoon correspondence construction approach with iterative graph based transductive learning (Graph-TL) and distance metric learning (DML) estimation and the local shape descriptor for cartoon application, which can successfully achieve rotation and scale invariance in cartoon matching.
- Short Text Classification for Faults Information of Secondary Equipment Based on Convolutional Neural Networks21
An automatic text classification based on convolutional neural networks (CNN) is proposed in this paper, and the validity and accuracy of the model is verified using actual operational data from the northwest power grid as case study.
- Towards Query Efficient Black-box Attacks21
Through extensive experiments, it is shown that with only 1,701 queries, an adversarial example can perturb a gray image to any target class of ImageNet with a 100% success rate on InceptionV3 and the algorithm has successfully defeated two real-world systems.
- Associations Between Sex-Specific Reproductive Factors and Risk of New-Onset Lung Cancer Among Female Patients20
Early age at menopause, a shortened reproductive life span, and early age at first birth were associated with higher risks of lung cancer, particularly of LUAD in a subpopulation with higher genetic susceptibility and detrimental behaviors.
- 20
- HAR-sEMG: A Dataset for Human Activity Recognition on Lower-Limb sEMG20
A novel dataset for HAR on lower-limb sEMG named “HAR-sEMG,” using 6 s EMG signal sensors attached to the left leg is established, and a benchmark summarizing experiments with many combinations of existing high-dimensional signal processing algorithms-based manifold learning on the dataset is provided for a performance analysis.
- 20
- UAV swarm formation reconfiguration control based on variable-stepsize MPC-APCMPIO algorithm19
The basic pigeon-inspired optimization (PIO) algorithm is introduced, and the Cauchy mutation operator, adaptive weight factor, and roulette wheel selection into the basic PIO are introduced, which is used to optimally solve the swarm objective functions.
- 19
- Multi-Level Graph Encoding with Structural-Collaborative Relation Learning for Skeleton-Based Person Re-Identification19
This work constructs graphs to model human skeletons from different levels, and for the first time proposes a Multi-level Graph encoding approach with Structural-Collaborative Relation learning (MG-SCR) to encode discriminative graph features for person Re-ID.
- 19
- 19
- LF-EME: Local features with elastic manifold embedding for human action recognition19
This work proposes a framework based on Elastic Manifold Embedding (EME), a new sparse manifold learning algorithm, together with local interest point features to handle human action recognition and the result is very promising in comparison with state-of-the-art methods.
- A reinforcement learning-based predator-prey model18
This work uses reinforcement learning algorithms to improve the existing individual-based ecosystem simulation algorithms, which allows species to spontaneously adjust their strategies according to a short period of experience and then feed back to improve their abilities to make action decisions.
- MeDJ: Multidimensional Emotion-aware Music Delivery for Adolescent18
The proposed MeDJ is a multidimensional emotion-aware music delivery application, which turns adolescents' music listening into a health and pleasure way, and aims at helping adolescents to improve emotional management and prevent depression.
- 18
- 18
- 17
- Clustering Multivariate Time Series Data via Multi-Nonnegative Matrix Factorization in Multi-Relational Networks17
A novel multivariate time series clustering method using multi-nonnegative matrix factorization (MNMF) in multi-relational networks to trace the evolutionary trends of clusters, thereby extending MNMF to dynamic MNMF.
- Support vector machine active learning by Hessian regularization17
A semi-supervised algorithm is introduced to learn the classifier and then an active learning scheme is performed on top of the semi- supervised scheme to achieve a better performance than the traditional active learning methods.
- Dimensionality reduction of data sequences for human activity recognition17
The framework of manifold elastic net is introduced that encodes the local geometry to find an aligned coordinate system for data representation and is efficient because classification error minimization criterion is utilized to directly link the classification error with the selected subspace.
- Data-driven facial animation via semi-supervised local patch alignment17
A novel data-driven facial animation technique which drives a neutral source face to get the expressive target face using a semi-supervised local patch alignment framework which demonstrates its superiority to existing methods.
- High-resolution digital holography with the aid of coherent diffraction imaging17
A method is proposed to enhance the resolution of digital holography in all directions by placing a random phase plate between the specimen and the electronic camera and then using an iterative approach to do the reconstruction.
- Multi-Camera Tracking Helmet System17
A novel multiple camera tracking helmet system is presented, which stitch views from multiple cameras mounted on the helmet to one wide view in which to tracking objects, this simplifies the task of object tracking from several cameras.
- 17
- Latent Degradation Representation Constraint for Single Image Deraining16
A novel Latent Degradation Representation Constraint Network (LDRCNet) that consists of the Direction-Aware Encoder, Deraining Network, and Multi-Scale Interaction Block is proposed to extract latent degradation representation adaptively by first using the deformable convolutions to exploit the direction property of rain streaks.
- Distill-then-prune: An Efficient Compression Framework for Real-time Stereo Matching Network on Edge Devices16
This paper proposes a novel strategy by incorporating knowledge distillation and model pruning to overcome the inherent trade-off between speed and accuracy and obtained a model that maintains real-time performance while delivering high accuracy on edge devices.
- 16
- HASA: Hybrid Architecture Search with Aggregation Strategy for Echinococcosis Classification and Ovary Segmentation in Ultrasound Images16
A hybrid NAS framework for ultrasound (US) image classification and segmentation using a pre-trained backbone and several searched cells, which takes advantage of the strengths of both NAS and the expert knowledge from existing convolutional neural networks.
- Instance-Level Embedding Adaptation for Few-Shot Learning16
A novel instance-level embedding adaptation mechanism, aiming at rapidly adapting embedding deep features to improve their generalization ability in recognizing novel categories, and achieves the-state-of-the-art performance.
- Single image super-resolution via subspace projection and neighbor embedding16
A novel learning-based single image super-resolution algorithm to address the problems of inefficient learning and improper estimation in coping with nonlinear high-dimensional feature data and applies iterative back projection (IBP) to further enhance the super- resolution results.
- 16
- Semantic segmentation of large-scale point clouds based on dilated nearest neighbors graph15
Instead of down-sampling point cloud directly, a dilated nearest neighbor encoding module is proposed to broaden the network’s receptive field to learn more 3D geometric information and without increase of network parameters, the method is computation and memory efficient for large-scale point clouds.
- 15
- 15
- Three-dimensional reconstruction of wafer solder bumps using binary pattern projection15
A new 3D reconstruction mechanism for inspecting the surface of solder bumps grown on wafer bumps is proposed, still based upon the light pattern projection framework, but uses the Ronchi pattern - a pattern that contrasts with the traditionally used gray level one.
- Associations of polycyclic aromatic hydrocarbons, water-soluble ions and metals in PM2.5 with liver function: Evidence from schizophrenia cohort14
Several PAHs were significantly associated with liver enzymes, while water-soluble ions and metal components had almost no association, which highlight the short-term combined effects of PM2.5 components on liver function in schizophrenia patients.
- 14
- 14
- Realtime and robust object matching with a large number of templates14
This work proposes a compact DOT representation with a fast partial occlusion handling approach, and introduces the C-DOT similarity map to store the matching scores of individual grids in each sliding window, which is used to further infer the Occlusion map.
- Interactive cartoon reusing by transfer learning14
Three visual features are defined: Hausdorff contour feature, color histogram and motion feature, to characterize the shape, color and motion structure information of a cartoon character and a new dimension reduction tool, termed sparse transfer learning (SPA-TL), is adopted to effectively and efficiently encode user's search intention.
- Characteristics and prognostic impact of cancer cachexia defined by the Asian Working Group for Cachexia consensus in patients with curable gastric cancer13
The Asian Working Group for Cachexia (AWGC) criteria is more effective in diagnosing cancer cachexia in the Asian population and provide better prognostic indicators.
- 13
- 13
- Data association and loop closure in semantic dynamic SLAM using the table retrieval method13
This paper proposes a table retrieval method for SLAM data association and loop closure using semantic information in a dynamic environment and achieves considerable performance improvement compared with state-of-the-art methods.
- 13
- 13
- Attention-aware concentrated network for saliency prediction13
Experimental results on seven public datasets demonstrate that the proposed ACNet outperforms the state-of-the-art models without any prior knowledge or post-processing.
- 13
- 13
- 12
- 12
- 12
- Spatial-Temporal Information Aggregation and Cross-Modality Interactive Learning for RGB-D-Based Human Action Recognition12
A novel approach for RGB-D human action recognition by aggregating spatial-temporal information and implementing cross-modality interactive learning, which denotes that the approach has advantages in recognizing subtle actions.
- Point‐selection and multi‐level‐point‐feature fusion‐based 3D point cloud classification12
The authors propose the point-selection structure which can calculate the importance of each point's feature and multi-level-point-feature fusion module is proposed to extract the features of the point at all levels and fuse them as a new feature of that point.
- A tensor framework for geosensor data forecasting of significant societal events12
A tensor pattern is used to model the geosensor data, based on which a tensor decomposition algorithm is then developed to estimate future values of geos sensor data, and a rank increasing strategy and a sliding window strategy are used to improve the prediction accuracy.
- Real-time 3D face modeling based on 3D face imaging12
A new real time 3D face modeling system was set up, which could generate real time3D face models with high quality by using a depth camera (such as Kinect) even the background of face imaging was complicated.
- Conditional simultaneous localization and mapping: A robust visual SLAM system12
A novel VSLAM approach called conditional simultaneous localization and mapping (C-SLAM) is proposed in which camera state transition is derived from image data using optical flow constraints and epipolar geometry in the prediction stage, which improves prediction accuracy and replaces commonly used predefined dynamic models which require additional computation.
- Biview Learning for Human Posture Segmentation from 3D Points Cloud12
Experimental results show that the proposed biview learning scheme significantly outperforms the state-of-the-art method for human posture segmentation.
- 12
- 12
- 11
- Deep-Reinforcement Learning-Based Co-Evolution in a Predator–Prey System11
This research showed that the predators’ reinforcement learning ability contributed to the stability of the ecosystem and helped predators obtain a more reasonable behavior pattern of coexistence with its prey and demonstrates the feasibility of predicting the behavior evolution in a predator–prey ecosystem using AI approaches.
- 11
- 11
- Residential greenness, air pollution, genetic predisposition and the risk of lung cancer10
Increased residential greenness was linearly associated with reduced risk of lung cancer, particularly adenocarcinoma, and such an association may be substantially mediated by mitigating air pollutants, especially PM2.5 and increased physical activity.
- Debiased Distillation for Consistency Regularization10
This work proposes a simple yet effective Intra-class Knowledge Distillation (IKD) method that facilitates knowledge sharing within the same class to ensure consistent predictions and reduces prediction bias by ensuring intra-class consistency.
- Pediatric Reference Change Value Optimized for Acute Kidney Injury: Multicenter Retrospective Study in China10
In this retrospective analysis of AKI criteria in PICU admissions in China, the pROCK criteria for AKI may have better utility in critically ill children, and is better correlated with severity and outcome ofAKI.
- 10
- 10
- Structured light-based shape measurement system10
By considering laser stripe distortion assumption, efficient belief propagation algorithm is adopted to extract center of laser stripe, which proves superior to existing peak detection approaches and called labels in machine learning.
- Unsupervised-clustering-driven noise-residue filter for phase images10
The proposed unsupervised-clustering-driven noise-residue filter is based on the fact that most residues are present in the form of adjacency caused by noisy wrapped phases, and reduces residue reduction, fringe preservation, and rapidity and improves the operating speed of phase unwrapping significantly.
- Quasispecies groups in the core promoter region of hepatitis B virus.10
The sequencing results indicate that there are HBV quasispecies groups in patients with chronic HBV infection, and there is a hot deletion region in the BCP.
- 9
- CbwLoss: Constrained Bidirectional Weighted Loss for Self-Supervised Learning of Depth and Pose9
The proposed method outperforms existing state-of-the-art self-supervised methods under the same conditions and without introducing additional auxiliary information, and mitigate the effect of textureless regions on model optimization by measuring the differences between features with more semantic and contextual information without requiring additional networks.
- HQDec: Self-Supervised Monocular Depth Estimation Based on a High-Quality Decoder9
This work proposes a high-quality decoder (HQDec), with which multilevel near-lossless fine-grained information, obtained by the proposed adaptive axial-normalized position-embedded channel attention sampling module (AdaAxialNPCAS), can be adaptively incorporated into a LR feature map with high-level semantics utilizing the proposed adaptive information exchange scheme.
- Acceptability of Tuberculosis Preventive Treatment Strategies Among Healthcare Workers Using an Online Survey - China, 2021.9
To further promote and implement TPT in China, practical measures included policy support, high-quality training for healthcare workers, and enhanced public awareness of TB prevention and control.
- 9
- Online tracking based on efficient transductive learning with sample matching costs9
An efficient transductive learning for online tracking is proposed by utilizing the correspondences among the generated unlabeled and labeled samples, modeled by a matching costs function to achieve more efficient learning of representative separators.
- 9
- 9
- Near‐infrared spectroscopy coupled with Gramian angular field two‐dimensional convolutional neural network for white tea adulteration detection8
Quantitative adulteration testing for tea origin has encountered challenges due to the poor performance in random external validation, which has severely hindered the practical application of near‐infrared (NIR) technology.
- 8
- Serum IL-12p40: A novel biomarker for early prediction of minimal change disease relapse following glucocorticoids therapy8
IL-12p40 may serve as a new biomarker for predicting the risk of MCD recurrence after glucocorticoid treatment, and it may be involved in the pathogenesis and recurrence of M CD.
- 8
- Online learning using projections onto shrinkage closed balls for adaptive brain-computer interface8
The effectiveness of the proposed adaptive projected sub-gradient method for adaptive EEG classification is revealed, making it a promising tool for on-device training and updating of wearable/portable BCIs, as well as for application in other related fields, such as EEG-based biometrics.
- 8
- 7
- 7
- 7
- 7
- 7
- Dynamic Parameters Identification Method of 6-DOF Industrial Robot Based on Quaternion7
An identification algorithm based on the least-squares method is proposed, which can realize the accurate identification of dynamic parameters of a 6-degrees-of-freedom (DOF) robot by establishing a dynamic model.
- A Graph Theory-Based Method for Dynamic Modeling and Parameter Identification of 6-DOF Industrial Robots7
The experiments prove that the result of parameter identification has extremely high accuracy and the dynamic model can accurately describe the 6R robot system mathematically.
- 7
- Mobile crowdsourcing based context-aware smart alarm sound for smart living7
A smart alarm sound recommendation system working with smartphones to deliver smart alarm sounds by considering not only specified information, such as sleep patterns, but also context information such as weather, and social information to provide a smart life.
- Recovering the absolute phase maps of three selected spatial-frequency fringes with multi-color channels7
This paper proposes a method to capture the fringe images of three spatial frequencies with multi-color channels and 3CCD camera and shows how the cross talk among color channels can be eliminated effectively and the measured object can be reconstructed with high accuracy.
- 7
- OPTIMIZATION OF BIT-PAIRING CODIFICATION WITH LEARNING FOR 3D RECONSTRUCTION7
This paper addresses how the spatial shifting of bit-pairing mechanism can be designed optimally for minimizing this number for faster inspection speed and proposes an optimal solution to shifting strategy optimization.
- A LEARNING BASED APPROACH TO AUDIO SURVEILLANCE IN HOUSEHOLD ENVIRONMENT7
This paper proposes a learning based approach to intelligent audio surveillance in household environment which can be applied to static or mobile audio surveillance systems which work solely or work as the supplements of video surveillance systems.
- 7
- 6
- 6
- 6
- PSMD-SLAM: Panoptic Segmentation-Aided Multi-Sensor Fusion Simultaneous Localization and Mapping in Dynamic Scenes6
This paper introduces PSMD-SLAM, a novel panoptic segmentation assisted multi-sensor fusion SLAM approach tailored for dynamic environments, which outperforms other SLAM algorithms in terms of accuracy and robustness, especially in dynamic environments.
- Hierarchical classified storage and incentive consensus scheme for building IoT under blockchain6
This work proposes a novel hierarchical message aggregation scheme and a classified storage method under Reed–Solomon (RS) coding to minimize storage overhead while ensuring data recoverability and query performance and introduces a dynamic incentive reputation consensus mechanism to tackle scalability challenges.
- 6
- 6
- Model-Based Human Pose Estimation with Hierarchical ICP from Single Depth Images6
Human poses which are presented by real 3D points cloud got from Kinect sensor are estimated and tracked by a hierarchical human model in ICP framework to solve several specific problems, such as leg- or arm-crossing, etc.
- 6
- Scaling behaviors of CG clusters for chromosomes6
A new method to study the scaling behaviors of CG clusters in different organism chromosomes is adopted, and the values of γ have the same behavior as thevalues of α in statistical distributions of P(S) ∝ e−αS.
- 5
- 5
- 5
- VRR-Net: Learning Vehicle–Road Relationships for Vehicle Trajectory Prediction on Highways5
A deep learning network that extracts features from vehicle–road relationships and models the effects of traffic environments on vehicles to achieve performance that is comparable to or better than that of state-of-the-art methods on the next-generation simulation dataset is presented.
- 5
- 5
- Joint learning of convolution neural networks for RGB‐D‐based human action recognition5
A single two-stream ConvNets framework for multimodality learning that extract features through RGB and depth streams is proposed and a features fusion and joint training strategy is adapted to learn RGB-D complementary features simultaneously.
- 5
- 5
- Guest Editorial Special Issue on Advancing Intelligent Automation in Sharing Economy5
A host of enabling technologies has reached the mainstream for the rise of sharing economy, including open data, the ubiquity of low-cost mobile phones, and social media, which dramatically reduce the friction of share-based business and organizational models.
- Efficient construction of topological semantic map with 3D information5
A novel framework for topological semantic map is constructed by projecting 3D scene semantic information recognized by convolutional neural network onto a 2D plane and semantic annotation of the intersection in the topological map is utilized to build up a complete object-based semantic map.
- 5
- A robust projector–camera interactive display system based on finger touch control by fusing finger and its shadow5
A robust projector–camera interactive display system consisting of one camera and one projector that enables users to interact with computer by touching on arbitrary surfaces with bare hand is proposed, which is robust to different lighting conditions.
- 5
- 5
- 5
- Semi-automatic cartoon generation by motion planning5
A semi-supervised graph-based cartoon reusing approach that aims at generating smooth cartoons from the existing data by calculating the distance based on local shape context, which is expected to be rotation and scaling invariant.
- Statistical properties of nucleotide clusters in DNA sequences5
Using the complete genome of Plasmodium falciparum 3D7 as an example, the distribution functions for the amount of C or G and A or T consecutively and non-overlapping blocks of m bases are examined to provide some insight into the nucleotide clusters of DNA sequences.
- 4
- 4
- 4
- 4
- Skeletal Muscle Mass Measurement Using Cone-Beam Computed Tomography in Patients With Head and Neck Cancer4
CBCT may be used as an alternative for CT to measure SMA in HNC patients during radiotherapy and demonstrated excellent interobserver reliability.
- 4
- Weighted ensemble networks for multiview based tiny object quality assessment4
A novel system for automatic TOQA that captures images of tiny object from multiple (more than two) view points is developed, and a novel method termed weighted ensemble network (WENet) to effectively integrate the information of different views is proposed.
- Face Translation based on Semantic Style Transfer and Rendering from One Single Image4
One face translation framework for translating human faces to that with visual effects from one single prototype image, which can generate images with prototype's visual effects while preserving the original person's identification and expression information is presented.
- View Invariant Human Action Recognition Using 3D Geometric Features4
This paper presents a straightforward and efficient approach for 3D human action recognition based on skeleton sequences that demonstrates the effectiveness of the method compared to the current state-of-the-art in cross-view evaluation, and significant improvement inCross-subjects evaluation.
- Driving event-related potential-based speller by localized posterior activities: An offline study4
Results showed that with visual graphic stimuli, localized posterior activities were enough to drive an ERP-based speller to achieve comparable or even better performance, compared to the traditional paradigm using global activities.
- 4
- 4
- 4
- 4
- 4
- 4
- Scaling behavior of nucleotide cluster in DNA sequences4
The relatively random walk length function L(n) and the local compositional complexity of nucleotide sequences based on a new model may provide some insight into nucleotide cluster of DNA sequence.
- 3
- 3
- 3
- Rivaroxaban plus aspirin vs. dual antiplatelet therapy in endovascular treatment in peripheral artery disease and analysis of medication utilization of different lesioned vascular regions3
Rivaroxaban plus aspirin improved outcomes compared to DAPT in patients with lower extremity artery disease and no significant difference was observed in the occurrence of primary efficacy events between the two groups.
- 3
- 3
- 3
- 3
- 3
- MOS: A Low Latency and Lightweight Framework for Face Detection, Landmark Localization, and Head Pose Estimation3
Online feedback sampling is proposed to augment the training samples across different scales, which increases the diversity of training data automatically and achieves the state-of-the-art performance in low computational resources.
- 3
- 3
- 3
- 3
- RETRACTED ARTICLE: Cross-camera multi-person tracking by leveraging fast graph mining algorithm3
A new online learning tracking system to generate the complete trajectory for each tracking object by applying linear programming and greedy algorithm to handle the data association problem to generate the complete trajectory for each tracking object.
- 3
- A Crowdsensing-Based Real-Time System for Finger Interactions in Intelligent Transport System3
This paper proposes a real-time projector-camera finger system based on the crowdsensing, in which user can interact with a computer by bare hand touching on arbitrary surfaces, and it can be used as an intelligent device in intelligent transport system where the driver can watch and interact with the display information while driving, without causing visual distractions.
- 3
- 3
- A Particle Swarm Optimizer Mimicking Parasitic Immune Behavior3
In this paper, in order to improve the searching ability of the algorithm, the particle swarm was divided into two populations called as parasitic and host, and elitist learning strategy is applied to the particle of parasite population to avoid the elite particles trapped in the local optimum.
- A Real-Time Dynamic Gesture Recognition System3
A fast and effective method is presented for hand detection and tracking at first for the trajectory extraction and a novel trajectory correction method is applied for simply but effectively trajectory correction.
- 3
- 3
- 3
- 3
- 2
- 2
- UCS: A Universal Model for Curvilinear Structure Segmentation2
The Universal Curvilinear structure Segmentation (UCS) model is proposed, which adapts SAM to CSS tasks while further enhancing its cross-domain generalization, and demonstrates state-of-the-art generalization and open-set segmentation performance across medical, engineering, natural, and plant imagery, establishing a new benchmark for universal CSS.
- 2
- Enhancing Incomplete Multi-modal Brain Tumor Segmentation with Intra-modal Asymmetry and Inter-modal Dependency2
A novel approach that enhances the BTS model from two perspectives is proposed, introducing a pre-training stage that generates a diverse pre-training dataset covering a wide range of different combinations of tumor shapes and brain anatomy and a post-training stage that enables the model to reconstruct missing modalities in the prediction results when only partial modalities are available.
- 2
- 2
- 2
- 2
- 2
- 2
- Semantic-Aware Feature Aggregation for Few-Shot Image Classification2
This work directly model the interdependencies between prototype features and image regions, resulting in a novel Semantic-Aware Feature Aggregation (SAFA) framework that can place more weights on category-relevant image regions.
- 2
- 2
- 2
- Characterization of light chain c-terminal extension sequence variant in one bispecific antibody2
An aberrant peak near light chain (LC) was observed in the non-reduced capillary electrophoresis sodium dodecyl sulfate (nrCE-SDS) electrophoretogram during cell line development of one bispecific antibody (BsAb) product, and the detected mass was about 944 Da higher than LC.
- 2
- 2
- 2
- 2
- 2
- 2
- HARaaS: HAR as a service using wifi signal in IoT-enabled edge computing: poster abstract2
A WiFi-based activity recognition service is designed following the HARaaS architecture, and the lightweight machine learning and deep learning model are incorporated in the service for accurate activity recognition.
- Towards Intelligent Decision Making in Emotion-aware Applications2
The feasibility of the proposed intelligent emotion-aware system (IES), constructed to provide multi-dimensional online social community data collection and processing approaches for decision making, is demonstrated and the system performance is evaluated based on this application.
- 2
- Response: Optical Coherence Tomography Angiography of Optic Disc and Macula Vessel Density in Glaucoma and Healthy Eyes2
It is demonstrated that VD and perfusion density were decreased significantly in chronic hypertension and stage 4 hypertensive retinopathy patients compared with normal controls and a prospective study with a large sample size of patients in varying stages of glaucoma and controls is recommended.
- 2
- 2
- 2
- 2
- 2
- Cauchy estimator discriminant analysis for face recognition2
A novelty subspace learning algorithm for face recognition named Cauchy estimator discriminant analysis (CEDA) is proposed, based on the patch alignment framework (PAF) and Cauche estimator theory, and large errors caused by samples that are easy to confuse could be overcome.
- Design and Implementation of a Service Robot for Elders2
A prototype of a service robot for elders is described, which has the functions of autonomous mobility in the home environment, manipulation for small commodities, speech recognition and synthesis in specific backgrounds, and movement detection for elders’ safety and basic nursing ability.
- 2
- 2
- 2
- 2
- 2
- A defects detection system for the surfaces of stampings2
A new visual system for detecting defects on the surfaces of stampings using a set of LED bar lights to enhance the imaging of the defects and punching edges in the vertical orientation of the surface, which facilitates the location of the defect such as scratch and pitting and the measurement of the punching sizes.
- 2
- 2
- 2
- 2
- 2
- A novel design of grating projecting system for 3D reconstruction of wafer bumps2
This paper describes a design consisting of a grating and several pieces of spherical lens, that addresses the requirements for 3D reconstruction with high speed and accuracy and can project a high spatial-frequency binary grating onto the inspected surface properly.
- 2
- [Construction and expression of humanized anti-HBsAg scFv targeting interferon-alpha in escherichia coli].2
The construction and expression of the fusion gene of humanized anti-HBsAg scFv and interferon in E.coli are successful and the fusion protein retained its specific binding activity and antiviral activities.
- Toward a distributed terabyte text retrieval system in China-US million book digital library2
A distributed terabyte text retrieval system for China-US Million Book Digital Library is developed, which can search less data while maintaining acceptable retrieval accuracy, and some new retrieval models and text clustering approaches are explored.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- CUCN: continuously updated connection network for low-light image enhancement1
The results show that the proposed CUCN method is superior to other state-of-the-art methods in terms of both subjective and objective metrics.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Neural Network Based Electronics Segmentation1
Experimental results show that the proposed framework has high recognition accuracy and strong robustness, which will be helpful when applied in robotic arm settings in automated assembly lines.
- 1
- 1
- Co-consistent Regularization with Discriminative Feature for Zero-Shot Learning1
An end-to-end framework is proposed, which is different from traditional ZSL methods in the following two aspects: (1) it uses a cascaded network to automatically locate discriminative regions, which can better extract latent features and contribute to the representation of key semantic attributes and (2) it achieves mapping in visual-semantic embedding space and calculation procedure of the dot product in deep learning framework.
- Sequential Probability Ratio Testing with Power Projective Base Method Improves Decision-Making for BCI1
The proposed SPRT method provides an explicit relationship between stopping time, thresholds, and error, which is important for balancing the time-accuracy trade-off, and suggest SPRT would be useful in speeding up decision-making while trading off errors in BCI.
- 1
- 1
- 1
- 1
- An Automatic Timestamp Replanting Algorithm for Panorama Video Surveillance1
A novel localization procedure which first localizes second-digit by using a pixel secondly-periodicity method and then it localizes timestamp via extracting all digits of timestamp and adopts a homography-based method to conduct timestamp removal.
- Research on Data Management for China Railway Based on Computer Science1
The challenges and opportunities with Big Data for application of China railway are summarized and it is pointed out there is more than enough that the authors can work on.
- 1
- 1
- 1
- 1
- A Novel Projector-Camera Interaction System with the Fingertip1
A novel algorithm is proposed, which robustly detects fingertip and archives touch-operated controls with a high precision of few millimeters, and a simple and robust hand detection method which can work well in a wide variety of situations including dark and luminous ambient illumination.
- Transductive Cartoon Retrieval by Multiple Hypergraph Learning1
This paper builds multiple hypergraphs on cartoon characters based on their features, in which each vertex is a character, and an edge links to multiple vertices, and the distance estimation between characters is avoided and the high-order relationship among characters can be explored.
- 1
- Exploring e-government in Beijing1
The Municipal Office of Information in Beijing is responsible for managing its e- government programme and has been advancing the e-government programme through technology innovations in the past 10 years.
- 1
- Noise-Residue Filtering Based on Unsupervised Clustering for Phase Unwrapping1
An unsupervised clustering driven noise-residue filter is proposed, and applied as a preprocessing procedure of phase unwrapping of noise-induced phase inconsistencies, or residues in the wrapped phase map.
- 1
- 1
- 1
- Boundary detection of projected fringes on surface with inhomogeneous reflectance function1
This paper makes use of the spatial-temporal image volume over the target surface to tackle the issue of inhomogeneous reflectance function and shows that the observed intensity profile across the images of a fixed point has the same up-and-down profile of the orignal binary gratings, regardless of the reflectance on thetarget surface.
- 1
- 1
- A Cache-Based Distributed Terabyte Text Retrieval System in CADAL1
A cache-based, distributed terabyte text retrieval system, with fulltext retrieval, distributed computing and caching techniques, for the China-America Digital Academic Library project.
- [Screening and application of human-derived HBsAg bacteriophage single chain antibody in clinical diagnosis].1
The application of HBsAg specific ScFv in immunohistochemistry was successfully achieved and demonstrated that the human single chain Fv antibody has a specific combination character with hepatitis B surface antigen of different sources and paraffin-embedded patients tissue specimens, it did not react with normal liver tissue and HCV.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Mechanical thrombectomy for treatment of thromboembolic complication before coiling of the ruptured intracranial aneurysms–
Mechanical thrombectomy followed by coiling of the aneurysms is a safe and reasonable treatment regime, which may improve the prognosis of patients, and thromboembolic event beforeAneurysm coiling is rare.
- –
- –
- –
- –
- –
- D3L: Decomposition of 3D Rotation and Lift from 2D Joint to 3D for Human Mesh Recovery–
This work proposes a novel approach, Decomposition of 3D Rotation and Lift from 2D Joint to 3D mesh (D3L), which can leverage human pose estimation methods, and avoid pose errors introduced by shape estimation overfitting.
- –
- Structure Unbiased Adversarial Model for Medical Image Segmentation–
A new image to image translation problem to reduce structural gap, in addition to the typical intensity distribution gap, is reformulate and experimental results show that the proposed SUAM has the capability to transfer both intensity and structure distributions.
- –
- –
- –
- –
- –
- –
- –
- Development of a 3D environment perception system for unmanned emergency repair robots for oil and gas pipelines–
A 3D environment perception system is developed, a 2D LiDAR is assembled on a rotating unit driven by a stepper motor, and a costumed 3D laser scanner is mounted on the robot to sense the 3D outline of the environment of the accident site.
- –
- –
- –
- Research on Sales Forecasting Method for Transmission Parts of Customized Production–
The results show that the model has high prediction accuracy and can provide reference for enterprise decision-making and overcomes the problems of low prediction accuracy that may exist in the traditional hybrid model.
- –
- –
- –
- –
- –
- –
- Research on Trust Prediction and Decision in C2C–
Firstly bayesian networks are used to forecast the trust level of the merchants, then the prediction results are combined with the payoff matrix of both parties to help the consumers to make a decision.
- Human Body Recognition from Depth Image Based on Part Size and Position–
A human body recognition method from depth image based on part size and position features based onRandom forest classifiers are trained with different parameters to demonstrate the feasibility of proposed approach.
- –
- –
- –
- –
- –
- –
- –
- Education Measure in the Teaching Reform of Physiological Experiment and Scientific Research Ability of Students–
It is important the sdutent is the centre in the course of college teaching because high ratio experiment in synthesize or planned investigative must be arranged in physiological experiment course.
- –
- –
- Establishing the experiment system in relatively independence, deepening experiment teaching reforms–
- –
- –
- –
- Molecular biological roles of nitric oxide synthase in viral hepatitis–
It is found that nitric oxide (NO) is concerned with many physiological functions in liver cells and the pathological processes of a variety of liver diseases and has potential antiviral activity.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Epitope M apping of Hepatitis C Virus NS3 Antigen from a Phage Peptide Library by Using Imm obilized Specific M onocionai Antibody–
The result provides a new approach for HCV therapy and vaccine developm ent and one epitope was confirm ed as m imotope of HCV N S3.
- [Preparation of human single chain Fv antibody specifically against hepatitis C virus E2 antigen].–
ScFv, having a sutestantial affinity and specificity and being easy to prepare, is valuable in the detection of HCV E2 antigen.
- –
- [Synergetic transactivating functions of HCV core and truncated HBV middle surface protein].–
It is suggested that the two kinds of virus proteins have transactivating effect on SV40 early promoter/enhancer, and they act synergistically to explain the mechanisms of liver injury or tumorigenesis induced by HCV or/and HBV infection.
- –
- Communication and Diagnostic Protocol of Control System on Heavy duty Vehicle–
The network technology of control system and the diagnostic protocol SAE?J1708,SAE?J1587,SAe?J1922, ISO?9141,ISO?14230 used in the heavy duty vehicle are introduced.
- THE INTEGRATED DEVELOPMENT SYSTEM FOR THE VEHICLE ELECTRONIC CONTROL SYSTEM–
A new integrated development system for a vehicle electronic control system that consists of a simulation computation, real time hardware simulation, and a road test of a real vehicle, integrated onto a microcomputer.
- –
- –
- The Comparative Study on the Technological Performance of PEPSET and FURAN No bake Resin bonded Sand–
The experimental results show that the PEPSET resin bonded no bake is better in the technological performance, it will probably be an important development direction of resin bond no bake sand in aluminum casting.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.