A5034675261- Video Coding and Compression Technologies
- Advanced Data Compression Techniques
- Advanced Vision and Imaging
- Inertial Sensor and Navigation
- Indoor and Outdoor Localization Technologies
- Advanced Computational Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Robotics and Sensor-Based Localization
- Advanced MRI Techniques and Applications
- Image and Video Quality Assessment
- Geophysics and Sensor Technology
- GNSS positioning and interference
- Advanced Image Fusion Techniques
- Advanced MEMS and NEMS Technologies
- Sparse and Compressive Sensing Techniques
- Image Enhancement Techniques
- Advanced Image Processing Techniques
- Medical Imaging Techniques and Applications
- Advanced Adaptive Filtering Techniques
- Video Analysis and Summarization
- Target Tracking and Data Fusion in Sensor Networks
- Blind Source Separation Techniques
- Retinal Imaging and Analysis
- Advanced Neural Network Applications
- Advanced Data Storage Technologies
Tianjin University
2016-2025
ShanghaiTech University
2025
Central South University
2022-2025
Chongqing Three Gorges University
2008-2024
PLA Air Force Aviation University
2024
First Affiliated Hospital Zhejiang University
2024
People's Liberation Army Air Force
2024
Taizhou People's Hospital
2024
Nanjing Medical University
2024
Nantong University
2024
Every year, for ten years now, the IPIN competition has aimed at evaluating real-world indoor localisation systems by testing them in a realistic environment, with movement, using EvAAL framework. The provided unique overview of state-of-the-art systems, technologies, and methods positioning navigation purposes. Through fair comparison performance achieved each system, was able to identify most promising approaches pinpoint critical working conditions. In 2020, included 5 diverse off-site...
Alzheimer's Disease (AD) is a neurodegenerative disorder that causes continuous decline in cognitive functions and eventually results death. An early AD diagnosis important for taking active measures to slow its deterioration. Traditional diagnoses are usually based on clinical experience, which limited by several realistic factors. In this paper, we focus exploiting deep learning techniques diagnose eye-tracking behaviors. Visual attention, as typical behavior, of great value detecting...
This paper proposed three methods to compensate the temperature energy influence drift of MEMS vibration gyroscope, including radial basis function neural network (RBF NN), RBF NN based on genetic algorithm (GA), and GA with Kalman filter (KF). Three‐axis gyroscope (Gyro X, Gyro Y, Z) output data are compensated analyzed in this paper. The experimental results proved correctness these methods, is effectively. indicate that, after NN‐GA‐KF method compensation, bias instability Gyros Z...
Eye-tracking technology has become a powerful tool for biomedical-related applications due to its simplicity of operation and low requirements on patient language skills. This study aims use the machine-learning models deep-learning networks identify key features eye movements in Alzheimer's Disease (AD) under specific visual tasks, thereby facilitating computer-aided diagnosis AD. Firstly, three-dimensional (3D) visuospatial memory task is designed provide participants with stimuli while...
Abstract Few‐layered 2D materials are promising candidates to build highly integrated memristors. The interlayer coupling between two different in heterostructure is crucial for their band structure modulation. In this study, an Au/WS 2 /WSe /Au van der Waals memristor reported, the type‐II alignment formed by few‐layered WS and WSe . of tuned annealing at temperatures under argon atmosphere. switching ratio I–V cycle number annealed 350 °C increased 10 5 300, which 1000 times 6 that...
A broadband tunable orbital angular momentum (OAM) mode converter based on a helical long-period fiber grating (HLPFG) inscribed in conventional single-mode (SMF) is experimentally demonstrated. The proposed all-fiber OAM the core-cladding dual resonance near dispersion turning point (DTP). can operate with bandwidth of 303.9 nm @ −3 dB and 182.2 −10 dB, which is, as far we know, widest for SMF. Furthermore, be dynamically tuned within large dynamic range (>80 nm) by simply twisting...
Alzheimer's disease (AD) is a degenerative neurological that occurs in the elderly with typical symptoms of decline cognition, manifested by eye movement behaviors. The key to AD treatment requires early detection cognitive impairment, which relies on frequent medical screening. This article proposes an Internet Things (IoT) architecture constructed eye-tracker (ET) nodes and cloud-based diagnosis enabled machine learning (ML), can provide convenient screening oculomotor abnormalities...
This paper proposes a static-dynamic hybrid malware detecting scheme for Android applications. While the static analysis could be defeated by transformation technique sometimes and dynamic needs high complexity, suggested methods can automatically deliver an unknown App to or path according whether decompiled(its feature) which overcomes both weakness. The experimental results show that is effective as its detection accuracy achieve 93.33%-99.28%.
As one of the representative types user-generated contents (UGCs) in social platforms, micro-videos have been becoming popular our daily life. Although naturally exhibit multimodal features that are rich enough to support representation learning, complex correlations across modalities render valuable information difficult integrate. In this paper, we introduced a attentive network (MARNET) learn complete and robust representations benefit micro-video multi-label classification. To address...
Field Programmable Gate Array (FPGA) has become an excellent hardware accelerator solution for convolutional neural networks (CNN). Meanwhile, optimizing methods such as model compression have been proposed. As most CNN accelerators focus on dense networks, to solve the problem of difficult deployment due irregular we propose a method sparse in our work. The storage and coding format data obtained by block pruning is designed make it friendly implement FPGA. Besides, also efficient simple...
Recently, with the growing popularity of micro-videos, multi-label learning has attracted increasing attention due to its potential commercial value in different scenarios. However, existing methods place more emphasis on alignment between explicit semantics and visual features, while neglecting exploration interactions at fine-grained semantic levels. To address this problem, we propose a novel dual-domain aligned deep hierarchical matrix factorization (DADHMF) method for micro-video...
The paper proposes a novel prediction paradigm in image coding based on Convolutional Neural Networks (CNN). A deep neural network is designed to provide accurate pixel-wise causal neighbourhood. proposed CNN method trained the high-activity areas and it incorporated lossless compression system for high-resolution photographic images. uses CNN-based as well LOCO-I, whereby predictor selection performed using local entropy-based descriptor. errors are encoded CALIC-based reference codec....
Existing research on the 2D image-based 3D model retrieval task focuses learning transferable representations directly to narrow domain discrepancy. However, it is not easy achieve in practice due significant variations across two domains. In addition, some methods design a discriminator distinguish feature arising from source or target domains for learning, which will lead an unexpected deterioration of discriminability. To settle these problems, we propose jointly and discriminative...