A5038473009- Domain Adaptation and Few-Shot Learning
- Anomaly Detection Techniques and Applications
- Remote-Sensing Image Classification
- Digital Imaging for Blood Diseases
- Advanced SAR Imaging Techniques
- Geophysical Methods and Applications
- Advanced Image and Video Retrieval Techniques
- Imbalanced Data Classification Techniques
- Video Surveillance and Tracking Methods
- Retinal Imaging and Analysis
- Automated Road and Building Extraction
Xiamen University
2022-2024
High resolution range profile (HRRP) has attracted increasing attention in radar automatic target recognition (RATR). However, the target-aspect missing problem non-cooperative targets recognition, which is one of most challenging tasks RATR, received very few contributions recently. The proposed work motivated by a simple observation, i.e., as compared with HRRP signals interesting targets, sufficient unlabeled are much easier to acquire. these often neglected since there no label...
Imbalanced class distributions widely exist in real-world aerial images, which brings a significant challenge to scene classification due the undesirable bias toward majority classes as well overfitting for minority classes. Although similarity between different may be inconsistent, they can measured by mean of feature statistics. This motivates us transfer statistics having similar Specifically, based on observation that each follow Gaussian distribution, across would thus described The...
In real-world scenarios, aerial image datasets are generally class imbalanced, where the majority classes have rich samples, while minority only a few samples. Such imbalanced bring great challenges to scene recognition. this paper, we explore novel two-stage contrastive learning framework, which aims take care of representation and classifier learning, thereby boosting Specifically, in stage, design data augmentation policy improve potential according characteristics images. And employ...