A5100440822- Sparse and Compressive Sensing Techniques
- Optical measurement and interference techniques
- Medical Image Segmentation Techniques
- Advanced Data Compression Techniques
- Advanced Image Processing Techniques
- Coronary Interventions and Diagnostics
- Robot Manipulation and Learning
- Robotic Path Planning Algorithms
- Image and Object Detection Techniques
- Occupational Health and Safety Research
- Advanced Neural Network Applications
- Remote Sensing and LiDAR Applications
- Research studies in Vietnam
- Smart Agriculture and AI
- Peripheral Artery Disease Management
- Vascular Procedures and Complications
- Image and Signal Denoising Methods
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- Image Processing Techniques and Applications
- Metallurgical Processes and Thermodynamics
Sichuan University
2015-2024
Chengdu University
2024
Qingdao University of Science and Technology
2012
A balloon dilatation catheter is a vital tool in percutaneous transluminal angioplasty. Various factors, including the material used, influence ability of different types balloons to navigate through lesions during delivery.Thus far, numerical simulation studies comparing impacts materials on trackability catheters has been limited. This project seeks unveil underlying patterns more effectively by utilizing highly realistic balloon-folding method compare made from materials.Two materials,...
Depth estimation from monocular vision sensor is a fundamental problem in scene perception with wide industrial applications. Previous works tend to predict the depth based on high-level features obtained by convolutional neural networks (CNNs) or rely encoder–decoder frameworks of Transformers. However, they achieved less satisfactory results, especially around object contours. In this article, we propose Transformer-based contour-aware module recover aid enhanced Besides, develop cascaded...
Because of the general lack multi-level hospital management collaboration performance effectiveness research, this paper proposes a Synergy Entropy-House Quality (HoQ) Measurement Model by innovatively combining House measure model with Entropy computing principle. Triangular fuzzy functions are used to determine importance degree parameter each element which combined results from evaluation elements, arrive at comprehensive collaborative computation result for various ensuring objectivity....
Quantization is a promising method that reduces memory usage and computational intensity of Deep Neural Networks (DNNs), but it often leads to significant output error hinder model deployment. In this paper, we propose Bias Compensation (BC) minimize the error, thus realizing ultra-low-precision quantization without fine-tuning. Instead optimizing non-convex process as in most previous methods, proposed BC bypasses step directly quantizing by identifying bias vector for compensation. We have...
In this paper, a TV logo detection system is proposed based on the deep learning architecture for specific task. Training robust object detector typically requires large amount of manually annotated data, which time-consuming. To reduce cost, we construct in weakly-supervised framework, accomplished by localization network Region Proposal Network (RPN) and classification Fast RCNN. Based observed priors typical pictures video frames, data preparation processing are performed carrying out...