A5068470874- Advanced Image and Video Retrieval Techniques
- Bioinformatics and Genomic Networks
- Advanced Graph Neural Networks
- Complex Network Analysis Techniques
- Visual Attention and Saliency Detection
- Text and Document Classification Technologies
- Advanced Neural Network Applications
- Human Pose and Action Recognition
- Multimodal Machine Learning Applications
- Face and Expression Recognition
- Video Surveillance and Tracking Methods
- Brain Tumor Detection and Classification
- Medical Image Segmentation Techniques
- Machine Learning in Bioinformatics
- Computational Drug Discovery Methods
- Domain Adaptation and Few-Shot Learning
- Image Retrieval and Classification Techniques
- Gait Recognition and Analysis
- Anomaly Detection Techniques and Applications
- Functional Brain Connectivity Studies
- Image Enhancement Techniques
- Graph Theory and Algorithms
- Gene expression and cancer classification
- Video Analysis and Summarization
- Radiomics and Machine Learning in Medical Imaging
Anhui University
2015-2024
Anhui Special Equipment Inspection Institute
2016
RGBT tracking receives a surge of interest in the computer vision community, but this research field lacks large-scale and high-diversity benchmark dataset, which is essential for both training deep trackers comprehensive evaluation methods. To end, we present La rge- s cale H igh-diversity [Formula: see text]nchmark short-term R GBT (LasHeR) work. LasHeR consists 1224 visible thermal infrared video pairs with more than 730K frame total. Each pair spatially aligned manually annotated...
Abstract Community detection in complex network has become a vital step to understand the structure and dynamics of networks various fields. However, traditional node clustering relatively new proposed link methods have inherent drawbacks discover overlapping communities. Node is inadequate capture pervasive overlaps, while often criticized due high computational cost ambiguous definition So, community still formidable challenge. In this work, we propose algorithm based on decomposition,...
Nuclear fusion is one of the most promising ways for humans to obtain infinite energy. Currently, with rapid development artificial intelligence, mission nuclear has also entered a critical period its development. How let more people understand and join in research effective means accelerate implementation fusion. This paper proposes first large model field fusion, XiHeFusion, which obtained through supervised fine-tuning based on open-source Qwen2.5-14B. We have collected multi-source...
In semi-supervised medical image segmentation, the poor quality of unlabeled data and uncertainty in model's predictions lead to models that inevitably produce erroneous pseudo-labels. These errors accumulate throughout model training, thereby weakening performance. We found these pseudo-labels are typically concentrated high-uncertainty regions. Traditional methods improve performance by directly discarding regions, but this can also result neglecting potentially valuable training data. To...
Recently, Graph Neural Networks (GNNs) have gained widespread application in automatic brain network classification tasks, owing to their ability directly capture crucial information non-Euclidean structures. However, two primary challenges persist this domain. First, within the realm of clinical neuro-medicine, signals from cerebral regions are inevitably contaminated with noise stemming physiological or external factors. The construction networks heavily relies on set thresholds and...
This paper investigates how to perform robust image saliency detection by adaptively leveraging different source data. Given the aligned RGB-T pair, we learn representations for each modality using deep convolutional neural networks (CNNs) at scales, which can capture multiscale context features and rich semantic information inherited from previous CNNs trained on ImageNet Dataset. Then, employ fully connected network layer concatenate CNN features, infer map modality. For incorporating RGB...