A5100398969- Biometric Identification and Security
- Face recognition and analysis
- Quantum Information and Cryptography
- Advanced Vision and Imaging
- Video Surveillance and Tracking Methods
- Quantum Computing Algorithms and Architecture
- Advanced Neural Network Applications
- Human Pose and Action Recognition
- Quantum Mechanics and Applications
- Image Processing Techniques and Applications
- Forensic and Genetic Research
- Face and Expression Recognition
- Anomaly Detection Techniques and Applications
- Domain Adaptation and Few-Shot Learning
- Mobile Health and mHealth Applications
- Advanced Image and Video Retrieval Techniques
- Advanced Image Processing Techniques
- Distributed Sensor Networks and Detection Algorithms
- Opinion Dynamics and Social Influence
- Distributed Control Multi-Agent Systems
- Image Enhancement Techniques
- Machine Learning in Healthcare
- Game Theory and Applications
- Adversarial Robustness in Machine Learning
- Digital Mental Health Interventions
Beijing Anzhen Hospital
2021-2025
Institute of Automation
2004-2025
Chinese Academy of Sciences
2006-2025
North University of China
2025
University of Chinese Academy of Sciences
2019-2024
Capital Medical University
2021-2024
Institute of Acoustics
2023-2024
Tsinghua University
2023-2024
Xi'an Jiaotong University
2015-2024
Universität Hamburg
2024
Unlike other biometrics such as fingerprints and face, the distinct aspect of iris comes from randomly distributed features. This leads to its high reliability for personal identification, at same time, difficulty in effectively representing details an image. paper describes efficient algorithm recognition by characterizing key local variations. The basic idea is that sharp variation points, denoting appearing or vanishing important image structure, are utilized represent characteristics...
After the outbreak of severe acute respiratory syndrome (SARS) in world 2003, human coronaviruses (HCoVs) have been reported as pathogens that cause symptoms tract infections. Recently, a new emerged HCoV isolated from epithelium unexplained pneumonia patients Wuhan seafood market caused major disease and has named coronavirus 2 (SARS-CoV-2). This virus causes lung symptoms, leading to condition “coronavirus 2019” (COVID-19). The emergence SARS-CoV-2 SARS-CoV widespread fear concern...
To detect illegal copies of copyrighted images, recent copy detection methods mostly rely on the bag-of-visual-words (BOW) model, in which local features are quantized into visual words for image matching. However, both limited discriminability and BOW quantization errors will lead to many false matches, make it hard distinguish similar images from copies. Geometric consistency verification is a popular technology reducing but neglects global context information thus cannot solve this...
The low spatial resolution of light-field image poses significant difficulties in exploiting its advantage. To mitigate the dependency accurate depth or disparity information as priors for super-resolution, we propose an implicitly multi-scale fusion scheme to accumulate contextual from multiple scales super-resolution reconstruction. is then incorporated into bidirectional recurrent convolutional neural network, which aims iteratively model relations between horizontally vertically adjacent...
Iris images captured in non-cooperative environments often suffer from adverse noise, which challenges many existing iris segmentation methods. To address this problem, paper proposes a high-efficiency deep learning based approach, named IrisParseNet. Different previous CNN-based methods, only focus on predicting accurate masks by following popular semantic frameworks, the proposed approach is complete solution, i.e., mask and parameterized inner outer boundaries are jointly achieved...
Abstract An unprecedented Pd‐catalyzed regioselective activation of gem‐difluorinated cyclopropanes induced by CC bond cleavage is reported. It provides a general and efficient access to variety 2‐fluoroallylic amines, ethers, esters, alkylation products in high Z‐selectivity, which are important skeletons many biologically active molecules. In addition, the transformation represents first application as reaction partners transition‐metal‐catalyzed cross‐coupling reaction.
Abstract Gout is one of the most common types inflammatory arthritis, caused by deposition monosodium urate crystals in and around joints. Previous genome-wide association studies (GWASs) have identified many genetic loci associated with raised serum concentrations. However, hyperuricemia alone not sufficient for development gout arthritis. Here we conduct a multistage GWAS Han Chinese using 4,275 male patients 6,272 normal controls (1,255 cases 1,848 were genotyped), an additional 1,644...
As a reliable approach to human identification, iris recognition has received increasing attention in recent years. The most distinguishing feature of an image comes from the fine spatial changes structure. So pattern representation must characterize local intensity variations signals. However, measurements minutiae are easily affected by noise, such as occlusions eyelids and eyelashes, localization error, nonlinear deformations, etc. This greatly limits accuracy systems. In this paper,...
In this paper, we present a new method for face recognition using range data. The proposed is based on both global statistics of geometrical features and local correlative facial surfaces. Firstly, analyze the performances common representations by histograms matching. Secondly, propose to encode relationships between points their neighbors, which are demonstrated own great power represent intrinsic structure Finally, two kinds supposed be complementary some extent, combination them proven...
Objective Poor lifestyle represents a health risk factor and is the leading cause of morbidity chronic conditions. The impact poor can be significantly altered by individual's behavioral modification. Although there are abundant promotion applications tools, they still limited in providing tailored social support that goes beyond their predefined functionalities. In addition, virtual coaching approaches unable to handle user emotional needs. Our approach presents human–virtual agent mediated...
Federated learning opens a number of research opportunities due to its high communication efficiency in distributed training problems within star network. In this paper, we focus on improving the for fully decentralized federated (DFL) over graph, where algorithm performs local updates several iterations and then enables communications among nodes. such way, rounds exchanging common interest parameters can be saved significantly without loss optimality solutions. Multiple numerical...
One of the most often implied benefits high-dimensional (HD) quantum systems is to lead stronger forms correlations, featuring increased robustness noise. Here, we experimentally demonstrate n-setting linear HD steering criterion. We verify large violation inequalities without full-state tomography. The lower bound 2.24±0.01 in 11 dimensions, exceeding (V<2) two-setting criteria. Hence, a higher strength has been revealed. Moreover, method for enhancing noise increasing dimension,...
As an aperture-divided computational imaging system, microlens array (MLA) -based light field (LF) is playing increasingly important role in computer vision. the trade-off between spatial and angular resolutions, deep learning (DL) image super-resolution (SR) methods have been applied to enhance resolution. However, existing DL-based methods, depth-varying defocus not considered both dataset development algorithm design, which restricts many applications such as depth estimation object...
A dynamic nano-coordination photothermal hydrogel was successfully used as a highly adaptable and antibacterial wound dressing for infected skin injuries.
The poor immunogenicity of solid tumors limits the efficacy ofanti-programmed cell death protein 1 (anti-PD1)-based immune checkpoint blockade (ICB); thus, less than 30% patients with cancer exhibit a response. Currently, there is still lack effective strategies for improving tumor immunogenicity.The antitumor effect ultrasound-stimulated nanobubbles (USNBs) alone and in combination an anti-PD1 antibody was evaluated RM1 (prostate cancer), MC38 (colon cancer) B16 (melanoma) xenograft mouse...
The uncontrollable acquisition process limits the performance of iris recognition. In process, various inevitable factors, including eyes, devices, and environment, hinder recognition system from learning a discriminative identity representation. This leads to severe degradation. this paper, we explore uncertain factors propose uncertainty embedding (UE) uncertainty-guided curriculum (UGCL) mitigate influence factors. UE represents an image using probabilistic distribution rather than...
Recent advances in self-supervised models have led to effective pretrained speech representations downstream emotion recognition tasks. However, previous research has primarily focused on exploiting by simply adding a linear head top of the model, while overlooking design network. In this paper, we propose temporal shift module with integrate channel-wise information without introducing additional parameters or floating-point operations per second. By incorporating module, developed...
Importance Success rates of pulmonary vein isolation (PVI) are modest for persistent atrial fibrillation (AF). Additional linear ablation beyond PVI has not been proved superior to alone in randomized trials. Ethanol infusion the Marshall (EIVOM) facilitates at mitral isthmus and may lead improved effectiveness a strategy. Objective To determine whether with radiofrequency energy combined EIVOM added improves sinus rhythm maintenance compared patients AF. Design, Setting, Participants The...
Skeleton-based action recognition has recently made significant progress. However, data imbalance is still a great challenge in real-world scenarios. The performance of current algorithms declines sharply when training suffers from heavy class imbalance. imbalanced actually degrades the representations learned by these methods and becomes bottleneck for recognition. How to learn unbiased key long-tailed In this paper, we propose novel balanced representation learning method address problem...
Abstract Background Hypercontractile phenotype (HP) of the left ventricle (LV) is an actionable therapeutic target in patients with chronic coronary syndromes (CCS), but its prevalence and meaning heart failure preserved ejection fraction (HFpEF) remains unclear. Purpose To assess myocardial work correlates HP HFpEF. Methods In a prospective, observational, single-center, cross-sectional study, we recruited 62 healthy controls (55 ± 7 yrs, 36 males, 58.1%) 129 (58 8 70 54.3%) HFpEF referred...