A5100758186- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Multimodal Machine Learning Applications
- COVID-19 diagnosis using AI
- Machine Learning and Data Classification
- Machine Learning and ELM
- Face and Expression Recognition
- Image Processing Techniques and Applications
- Video Surveillance and Tracking Methods
- Image Retrieval and Classification Techniques
- Advanced Chemical Physics Studies
- Employment and Welfare Studies
- Cancer-related molecular mechanisms research
- Anomaly Detection Techniques and Applications
- Industrial Vision Systems and Defect Detection
- Migration, Ethnicity, and Economy
- Face recognition and analysis
- Energy Efficient Wireless Sensor Networks
- Neural Networks and Applications
- IoT-based Smart Home Systems
- Superconductivity in MgB2 and Alloys
- UAV Applications and Optimization
- Viral Infections and Outbreaks Research
- Random lasers and scattering media
Xidian University
2025
Ritsumeikan University
2022-2025
Environment and Plant Protection Research Institute
2025
Chinese Academy of Tropical Agricultural Sciences
2025
Lanzhou University of Technology
2017-2024
Beijing University of Posts and Telecommunications
2011-2024
Shandong University of Finance and Economics
2024
Shenyang Fire Research Institute
2024
Pacific Northwest National Laboratory
2023
Beijing Normal University
2021-2023
Key for solving fine-grained image categorization is finding discriminate and local regions that correspond to subtle visual traits. Great strides have been made, with complex networks designed specifically learn part-level feature representations. In this paper, we show it possible cultivate details without the need overly complicated network designs or training mechanisms -- a single loss all takes. The main trick lies how delve into individual channels early on, as opposed convention of...
Fine-grained vehicle classification is a challenging topic in computer vision due to the high intraclass variance and low interclass variance. Recently, considerable progress has been made fine-grained huge success of deep neural networks. Most studies based on networks, focus network structure improve performance. In contrast existing works classification, we loss function network. We add regularization term cross-entropy propose new function, Dual Cross-Entropy Loss. The places constraint...
Few-shot learning for fine-grained image classification has gained recent attention in computer vision. Among the approaches few-shot learning, due to simplicity and effectiveness, metric-based methods are favorably state-of-the-art on many tasks. Most of assume a single similarity measure thus obtain feature space. However, if samples can simultaneously be well classified via two distinct measures, within class distribute more compactly smaller space, producing discriminative maps....
Convolutional neural networks (CNNs) have recently shown excellent performance on the task of fine-grained vehicle classification, where motivation is to identify categories given vehicles. Generally speaking, main conventional back-propagation algorithm optimize loss function. The itself does not guarantee if extracted features are discriminative for classification. Intuitively, we can learn more with a relatively small number feature maps, generalization ability CNNs will be significantly...
Metric-based methods achieve promising performance on few-shot classification by learning clusters support samples and generating shared decision boundaries for query samples. However, existing ignore the inaccurate class center approximation introduced limited number of samples, which consequently leads to biased inference. Therefore, in this paper, we propose reduce error calibration. Specifically, introduce so-called Pair-wise Similarity Module (PSM) generate calibrated centers adapted...
Metric-based methods are one of the most common to solve problem few-shot image classification. However, traditional metric-based suffer from overfitting and local feature misalignment. The recently proposed reconstruction-based approach, which reconstructs query features support set a given class compares distance between original reconstructed as classification criterion, effectively solves misalignment problem. issue still has not been considered. To this end, we propose...
The main challenge for fine-grained few-shot image classification is to learn feature representations with higher inter-class and lower intra-class variations, a mere few labelled samples. Conventional learning methods however cannot be naively adopted this setting -- quick pilot study reveals that they in fact push the opposite (i.e., variations variations). To alleviate problem, prior works predominately use support set reconstruct query then utilize metric determine its category. Upon...
Few-shot fine-grained image classification has attracted considerable attention in recent years for its realistic setting to imitate how humans conduct recognition tasks. Metric-based few-shot classifiers have achieved high accuracies. However, their metric function usually requires two arguments of vectors, while transforming or reshaping three-dimensional feature maps vectors can result loss spatial information. Image reconstruction is thus involved retain more appearance details: the test...
The main challenge for fine-grained few-shot image classification is to learn feature representations with higher inter-class and lower intra-class variations, a mere few labelled samples. Conventional learning methods however cannot be naively adopted this setting - quick pilot study reveals that they in fact push the opposite (i.e., variations variations). To alleviate problem, prior works predominately use support set reconstruct query then utilize metric determine its category. Upon...
Litchi, an important tropical fruit, is severely affected by anthracnose disease. However, the mechanism of its disease resistance response remains unknown, and resistant accession genetic resources resistance-related genes have not yet been identified. In this study, 82 accessions litchi were evaluated for to Colletotrichum gloeosporioides, 'Haiken 5' 'Nongmei 5 hao' identified as susceptible, respectively. Leaves from these two inoculated with C. gloeosporioides collected at 6 24 h use...
Softmax cross-entropy loss with L2 regularization is commonly adopted in the machine learning and neural network community. Considering that traditional softmax simply focuses on fitting or classifying training data accurately but does not explicitly encourage a large decision margin for classification, some functions are proposed to improve generalization performance by solving problem. However, these enhance difficulty of model optimization. In addition, inspired regularized logistic...
A deep neural network of multiple nonlinear layers forms a large function space, which can easily lead to overfitting when it encounters small-sample data. To mitigate in classification, learning more discriminative features from data is becoming new trend. this end, paper aims find subspace networks that facilitate decision margin. Specifically, we propose the Orthogonal Softmax Layer (OSL), makes weight vectors classification layer remain orthogonal during both training and test processes....
Few-shot image classification is a challenging topic in pattern recognition and computer vision. fine-grained even more challenging, due to not only the few shots of labelled samples but also subtle differences distinguish subcategories images. A recent method called task discrepancy maximisation (TDM) can be embedded into feature map reconstruction network (FRN) generate discriminative features, by preserving appearance details through reconstructing query then assigning higher weights...
Considering that in neural network based on softmax cross entropy loss, the output probability is mainly linear computation of parameter vectors each class last layer and hidden features sample points. Therefore, final effected by L2-norm vector class. Taking binary-class as an example, if a has large L2-norm., decision boundary close to another with smaller so points will be easily assigned L2-norm. Based it, this paper proposes new which adjusts position it not biased any Experimental...