A5100689914- Advanced Neural Network Applications
- Radiomics and Machine Learning in Medical Imaging
- Human Pose and Action Recognition
- Gait Recognition and Analysis
- Video Surveillance and Tracking Methods
- Medical Image Segmentation Techniques
- Advanced Vision and Imaging
- Multimodal Machine Learning Applications
- Cloud Computing and Resource Management
- IoT and Edge/Fog Computing
- Topic Modeling
- Hand Gesture Recognition Systems
- Microbial bioremediation and biosurfactants
- Natural Language Processing Techniques
- Industrial Vision Systems and Defect Detection
- COVID-19 diagnosis using AI
- Advanced Image Processing Techniques
- Image and Video Stabilization
- Building Energy and Comfort Optimization
- Pancreatitis Pathology and Treatment
- Medical Imaging and Analysis
- Image Enhancement Techniques
- Diabetic Foot Ulcer Assessment and Management
- Web Data Mining and Analysis
- Visual Attention and Saliency Detection
Fudan University
1992-2025
Xuzhou Medical College
2025
First Affiliated Hospital of Shantou University Medical College
2025
Collaborative Innovation Center of Chemistry for Energy Materials
2025
Shantou University
2025
Edith Cowan University
2025
Lenovo (China)
2014-2024
Tianjin University
2023-2024
Robert Bosch (Germany)
2019-2024
Temper (United States)
2024
Gait recognition, applied to identify individual walking patterns in a long-distance, is one of the most promising video-based biometric technologies. At present, gait recognition methods take whole human body as unit establish spatio-temporal representations. However, we have observed that different parts possess evidently various visual appearances and movement during walking. In latest literature, employing partial features for description has been verified being beneficial recognition....
Purpose: Accurate segmentation of lung and infection in COVID-19 CT scans plays an important role the quantitative management patients. Most existing studies are based on large private annotated datasets that impractical to obtain from a single institution, especially when radiologists busy fighting coronavirus disease. Furthermore, it is hard compare current methods as they developed different datasets, trained settings, evaluated with metrics. Methods: To promote development data-efficient...
The emergence of deep learning has considerably advanced the state-of-the-art in cardiac magnetic resonance (CMR) segmentation. Many techniques have been proposed over last few years, bringing accuracy automated segmentation close to human performance. However, these models all too often trained and validated using imaging samples from single clinical centres or homogeneous protocols. This prevented development validation that are generalizable across different centres, conditions scanner...
Occluded person re-identification (Re-ID) focuses on addressing the occlusion problem when retrieving of interest across non-overlapping cameras. With increasing demand for intelligent video surveillance and application Re-ID technology, real-world draws considerable from researchers. Although a large number occluded methods have been proposed, there are few surveys that focus occlusion. To fill this gap help boost future research, article provides systematic survey Re-ID. In work, we review...
Congestive heart failure (CHF) is a serious pathophysiological condition with high morbidity and mortality, which hard to predict diagnose in early age. Artificial intelligence deep learning combining cardiac rhythms physiological time series provide potential help solving it. In this paper, we proposed novel method that combines convolutional neural network (CNN) distance distribution matrix (DDM) entropy calculation classify CHF patients from normal subjects, demonstrated the effectiveness...
Recent advances in vision-language pre-training have significantly enhanced the model capabilities on grounded object detection. However, these studies often pre-train with coarse-grained text prompts, such as plain category names and brief phrases. This limitation curtails model's capacity for fine-grained linguistic comprehension leads to a significant decline performance when faced detailed descriptions or contextual information. To tackle problems, we develop DoGA: Detect objects Grouped...
Two genes of the meta pathway phenol degradation were cloned from a phenol-utilizing strain Bacillus stearothermophilus and mapped by subcloning use Tn5 insertion mutation. They code for hydroxylase catechol 2,3-dioxygenase, respectively. The gene encoding which is more thermostable than 2,3-dioxygenase encoded other gene, shares rather limited homology with that Pseudomonas putida.
Automatic liver segmentation from abdominal Computed Tomography (CT) is an important step for hepatic disease diagnosis. It a challenging task owing to the similarity between and its adjacent organs low contrast of texture (e.g. tumors blood veins). In this paper, we propose cascaded structure automatically segment in CT scans. First, train fully convolutional neural network (FCN) coarse segmentation; second, make comparative study performance different classical models as post-processing...
Abstract Lithium-ion battery energy storage cabin has been widely used today. Due to the thermal characteristics of lithium-ion batteries, safety accidents like fire and explosion will happen under extreme conditions. Effective management can inhibit accumulation spread heat. This paper studies air cooling heat dissipation influence guide plate on cooling. Firstly, a simulation model is established according actual cabin, which divided into two types: with without plate. Then, at environment...
Sequence to sequence (seq2seq) prediction is a key many tasks of machine learning. Personal computer software sequence, as one these tasks, was regarded stochastic and unpredictable in the past. However, deep neural networks (DNNs) have achieved excellent performance recently especially field natural language process (NLP) such model, translation dialogue systems. This paper examines most popular DNNs approaches: LSTM, Encoder-Decoder network Memory handle learning task. Then three modified...