A5100332491- Anomaly Detection Techniques and Applications
- Time Series Analysis and Forecasting
- Gaze Tracking and Assistive Technology
- Advanced Vision and Imaging
- Advanced Computing and Algorithms
- Advanced Image Processing Techniques
- Traffic Prediction and Management Techniques
- Network Security and Intrusion Detection
- Authorship Attribution and Profiling
- Image and Video Quality Assessment
- Functional Brain Connectivity Studies
- Medical Image Segmentation Techniques
- Visual Attention and Saliency Detection
- Quantum Information and Cryptography
- Multimodal Machine Learning Applications
- Nanowire Synthesis and Applications
- Ga2O3 and related materials
- Ammonia Synthesis and Nitrogen Reduction
- Video Coding and Compression Technologies
- Robotics and Sensor-Based Localization
- Brain Tumor Detection and Classification
- Complex Systems and Time Series Analysis
- Adrenal Hormones and Disorders
- Advanced Neural Network Applications
- Advanced Image Fusion Techniques
Fuzhou University
2023
Nanjing University
2007-2022
East China University of Science and Technology
2022
Jiangsu University
2022
Collaborative Innovation Center of Advanced Microstructures
2022
Nanjing University of Aeronautics and Astronautics
2020
University of Jinan
2019
Wuhan National Laboratory for Optoelectronics
2013
Xijing University
2010
Hebei Normal University
2008
Human activity recognition (HAR) based on sensing data from wearable and mobile devices has become an active research area in ubiquitous computing, it envisions a wide range of application scenarios social networking, environmental context sensing, health well-being monitoring, etc. However, manually annotated is manpower-expensive, time-consuming, privacy-sensitive, which prevents HAR systems being really deployed scale. In this paper, we address the problem unsupervised human recognition,...
Anomaly detection on multivariate time series (MTS) is an important research topic in data mining, which has a wide range of applications information technology, financial management, manufacturing system, and so on. However, the state-of-the-art unsupervised deep learning models for MTS anomaly are vulnerable to noise have poor performance training containing anomalies. In this article, we propose novel Self-Training based Detection with Generative Adversarial Network (GAN) model called...
Edge detection is a fundamental technique in various computer vision tasks. Edges are indeed effectively delineated by pixel discontinuity and can offer reliable structural information even textureless areas. State-of-the-art heavily relies on pixel-wise annotations, which labor-intensive subject to inconsistencies when acquired manually. In this work, we propose novel self-supervised approach for edge that employs multi-level, multi-homography transfer annotations from synthetic real-world...
This paper proposes a real-time trajectory prediction method for quadrotors based on bidirectional gated recurrent unit model. Historical data of ten types were obtained. The units constructed and utilized to learn the historic data. results compared with traditional test its performance. efficiency proposed algorithm was investigated by comparing training loss time. over testing datasets showed that model produced better than baseline models all scenarios datasets. It also found can...
Multi-arm bandit (MAB) and stochastic linear (SLB) are important models in reinforcement learning, it is well-known that classical algorithms for bandits with time horizon T suffer from the regret of at least square root T. In this paper, we study MAB SLB quantum reward oracles propose both order polylog regrets, exponentially improving dependence terms To best our knowledge, first provable speedup regrets problems general exploitation learning. Compared to previous literature on exploration...
Purpose. A prospective controlled study was designed to observe the pharmacodynamics of rocuronium in cholestatic patients with or without hepatocellular injury. Methods. Sixty undergoing abdominal surgery were allocated into three groups: group I had 20 injury; II injury, and III (control group) hepatic disease. Anesthetized propofol fentanyl, all received 0.6 mg/kg for initial dose followed by intermittent repeated administration 0.15 mg/kg. The twitch high adductor pollicis muscle...
Multivariate time series (MTS) clustering is an important technique for discovering co-evolving patterns and interpreting group characteristics in many areas including economics, bioinformatics, data science, etc. Although has been widely studied the past decades, no enough attention paid to capture time-varying correlation MTS. In this article, we propose a novel approach MTS based on features. We introduce Gaussian Markov Random Fields (T-GMRF) model describe structure between variables,...
In this paper, an effective and efficient content-based video copy detection method is proposed. This based on temporal features of key frames. Firstly, each divided into shots, shot represented by its frame. Secondly, frame several sub-blocks, variations corresponding sub-blocks along series are extracted as fingerprint. Finally, fingerprints query target compared to determine whether they copies. Experimental results show that the proposed promising.
This work proposes to combine neural networks with the compositional hierarchy of human bodies for efficient and complete parsing. We formulate approach as a information fusion framework. Our model assembles from three inference processes over hierarchy: direct (directly predicting each part body using image information), bottom-up (assembling knowledge constituent parts), top-down (leveraging context parent nodes). The inferences explicitly decompositional relations in bodies, respectively....
Recent stereo matching methods, especially end-to-end deep networks, have achieved remarkable performance in the fields of autonomous driving and depth sensing. However, state-of-the-art algorithms, even with neural network framework, still difficulties at finding correct correspondences near-range regions object edge cues. To reinforce precision disparity prediction, present study, we propose a parallax attention algorithm based on improved group-wise correlation to learn content from...
Gaze object prediction is a newly proposed task that aims to discover the objects being stared at by humans. It of great application significance but still lacks unified solution framework. An intuitive incorporate an detection branch into existing gaze method. However, previous methods usually use two different networks extract features from scene image and head image, which would lead heavy network architecture prevent each joint optimization. In this paper, we build novel framework named...