A5100781388- Generative Adversarial Networks and Image Synthesis
- IoT and Edge/Fog Computing
- Cloud Computing and Resource Management
- Data Stream Mining Techniques
- AI in cancer detection
- Advanced Neural Network Applications
- Software System Performance and Reliability
- Infrastructure Maintenance and Monitoring
- Automated Road and Building Extraction
- Wildlife-Road Interactions and Conservation
- Machine Learning and Data Classification
- Advanced Image Processing Techniques
- Mobile Crowdsensing and Crowdsourcing
- Single-cell and spatial transcriptomics
- Image and Video Stabilization
- COVID-19 diagnosis using AI
- Scientific Computing and Data Management
- Auction Theory and Applications
- Retinal Imaging and Analysis
- Gene expression and cancer classification
- Distributed Sensor Networks and Detection Algorithms
- Data Management and Algorithms
- Maritime Navigation and Safety
- Colorectal Cancer Screening and Detection
- Music and Audio Processing
University of Science and Technology Liaoning
2024
University of Minnesota
2024
Inner Mongolia Agricultural University
2022
North China University of Technology
2019-2022
Namseoul University
2022
Qingdao Agricultural University
2022
Xiangtan University
2022
Institute of Computing Technology
2022
Chinese Academy of Sciences
2022
Purdue University West Lafayette
2021-2022
While the technology development towards microservices can significantly improve speed and agility of software service delivery, it also raises operational complexity associated with modern applications. This has led to emergence Service Mesh, a promising approach mitigate this situation by introducing dedicated infrastructure layer over without imposing modification on implementations. Aiming inspire more practical research work in exploited area, we paper present comprehensive review state...
With the rapid advancement of deep learning technologies, computer vision has shown immense potential in retail automation. This paper presents a novel self-checkout system for based on an improved YOLOv10 network, aimed at enhancing checkout efficiency and reducing labor costs. We propose targeted optimizations model, incorporating detection head structure from YOLOv8, which significantly improves product recognition accuracy. Additionally, we develop post-processing algorithm tailored...
In the past decade, commercial crowdsourcing platforms have revolutionized ways of classifying and annotating data, especially for large datasets. Obtaining labels a single instance can be inexpensive, but datasets, it is important to allocate budgets wisely. With limited budgets, requesters must trade-off between quantity labeled instances quality final results. Existing budget allocation methods achieve good cannot guarantee high individual under tight budget. However, in some scenarios,...
Utilizing multiple modalities to learn a good distance metric is of vital importance for various clinical applications. However, it common that are incomplete some patients due technical and practical reasons in healthcare datasets. Existing learning methods cannot directly the on such data with missing modalities. Nevertheless, contains valuable information characterize patient similarity modality relationships, they should not be ignored during process. To tackle aforementioned challenges,...
Linked data consist of both node attributes, e.g., Preferences, posts and degrees, links which describe the connections between nodes. They have been widely used to represent various network systems, such as social networks, biological networks etc. Knowledge discovery on linked is great importance many real applications. One major challenges learning how effectively efficiently extract useful information from attributes in data. Current studies this topic either use selected topological...
In this paper, we provide a quality of information (QoI) based data selection and transmission service for classification missions in sensor networks. We first identify the two aspects QoI, reliability redundancy, then propose metrics to estimate them. particular, implies degree which node contributes mission, can be estimated through exploring agreement between majority others. On other hand, redundancy represents overlap among different nodes, measured via investigating similarity their...
The past decades have witnessed significant progress towards improving the accuracy of predictions powered by complex machine learning models. Despite much success, lack model interpretability prevents usage these techniques in life-critical systems such as medical diagnosis and self-driving systems. Recently, issue has received attention, one critical task is to explain why a predictive makes specific decision. We refer this outcome interpretation. Many interpretation methods been developed...
Recent advances in spatial transcriptomics have enabled the comprehensive measurement of transcriptional profiles while retaining contextual information. Identifying domains is a critical step analysis spatially resolved transcriptomics. Existing unsupervised methods perform poorly on this task owing to large amount noise and dropout events transcriptomic profiles. To address problem, we first extend an algorithm supervised learning method that can identify useful features reduce hindrance....
Abstract Efficient and rapid auxiliary diagnosis of different grades lung adenocarcinoma is conducive to helping doctors accelerate individualized treatment processes, thus improving patient prognosis. Currently, there often a problem large intra-class differences small inter-class between pathological images tissues under grades. If attention mechanisms such as Coordinate Attention (CA) are directly used for grading tasks, it prone excessive compression feature information overlooking the...
As the number of service clusters in OpenStack Cloud Platform, work-load data center also increase, leading to node failures and performance issues. Therefore, managers need know how cloud platform is operating storing. This function can be realized through monitoring system, improve quality computing services help identify faults within system. The purpose this paper provide a solution for services, that allows users optimize resources based on changing business requirements First all,...
In the extensive age, dear designate perplexity and relatively supercilious show charge in traditive parcel extend project composition, double discriminator GAN is ply to bale work indicate composition. On basis of BicycleGAN, a topic added, analogous privation sine external province are reformed. proof, input aim likeness suit “margin idol + source cast,” product goal copy with top 10 chance genuineness. The trial arise appraised from three aspects variegation, PSNR importance, SSIM...