Export Citations
Save this search
Please login to be able to save your searches and receive alerts for new content matching your search criteria.
- research-articleJune 2024
FedRL: Federated Learning with Non-IID Data via Review Learning
ICMLC '24: Proceedings of the 2024 16th International Conference on Machine Learning and ComputingFebruary 2024, pp 115–120https://doi.org/10.1145/3651671.3651704Federated Learning epitomizes a sophisticated distributed machine learning methodology, enabling collaborative neural network model training across multiple entities without necessitating the transfer of local data, thereby fortifying data privacy ...
- research-articleJune 2024
Mpox-PyramidTransferNet: A Hierarchical Transfer Learning Framework for Monkeypox and Dermatological Disease Classification
ICMLC '24: Proceedings of the 2024 16th International Conference on Machine Learning and ComputingFebruary 2024, pp 404–410https://doi.org/10.1145/3651671.3651679During the COVID-19 pandemic, the global immunity of human populations was adversely affected, posing threats for other contagious diseases. Monkeypox cases have surged recently, bringing new challenges to global public health. In this study, we propose ...
- research-articleMay 2024
Incorporating attribute-level aligned comparative network for generalized zero-shot learning
AbstractThe key challenge of zero-shot learning (ZSL) is to sufficiently disentangle each latent attribute from the class-level semantic annotations of images, thereby achieving a desirable semantic transfer to unseen classes with the disentangled ...
Highlights- Our IAAC-net is based on a novel incorporating attribute-level alignment strategy.
- C2C branch establishes diverse alignments to facilitate attribute disentanglement.
- CA loss is further proposed for accurate attribute ...
- research-articleMay 2024
Subdomain alignment based open-set domain adaptation image classification
Journal of Visual Communication and Image Representation (JVCIR), Volume 98, Issue CFeb 2024https://doi.org/10.1016/j.jvcir.2024.104047AbstractDomain adaptation has achieved great success in using labeled source domain samples to identify unlabeled target domain samples. Here, we aim to solve the open-set domain adaptation, which is different from the closed-set domain adaptation in ...
- research-articleMay 2024
Improving vision transformer for medical image classification via token-wise perturbation
Journal of Visual Communication and Image Representation (JVCIR), Volume 98, Issue CFeb 2024https://doi.org/10.1016/j.jvcir.2023.104022AbstractTransformer has achieved impressive successes for various computer vision tasks. However, most of existing studies require to pretrain the Transformer backbone on a large-scale labeled dataset (e.g., ImageNet) for achieving satisfactory ...
-
- research-articleMay 2024
Top-tuning: A study on transfer learning for an efficient alternative to fine tuning for image classification with fast kernel methods
Image and Vision Computing (IAVC), Volume 142, Issue CFeb 2024https://doi.org/10.1016/j.imavis.2023.104894AbstractThe impressive performance of deep learning architectures is associated with a massive increase in model complexity. Millions of parameters need to be tuned, with training and inference time scaling accordingly, together with energy consumption. ...
Highlights- A training-time saving transfer learning with pre-trained convolutional features.
- A fast kernel method with pre-trained features for image classification.
- A systematic empirical study comparing top-tuning and fine-tuning ...
- research-articleMay 2024
Extracting section structure from resumes in Brazilian Portuguese
Expert Systems with Applications: An International Journal (EXWA), Volume 242, Issue CMay 2024https://doi.org/10.1016/j.eswa.2023.122495AbstractThis paper presents a novel resume parser designed to effectively reorganize the textual content of any resume into its original section structure. Our work addresses two practical challenges overlooked by the existing literature: (i) ensuring ...
Highlights- A parser to retrieve the resume structure from a given pdf document.
- A simple approach to ensure the correct resume reading order.
- Two segmentation models to extract the sections and subsections from a resume.
- An anonymized ...
- research-articleMay 2024
Enhanced transfer learning with data augmentation
Engineering Applications of Artificial Intelligence (EAAI), Volume 129, Issue CMar 2024https://doi.org/10.1016/j.engappai.2023.107602AbstractTraditional machine learning methods require the assumption that training and test data are drawn from the same distribution, which proves challenging in real-world applications. Moreover, deep learning models require a substantial amount of ...
- research-articleMay 2024
A Comparative Study of Swin-Based Enhanced Remote Sensing Image Classifications
IPMV '24: Proceedings of the 2024 6th International Conference on Image Processing and Machine VisionJanuary 2024, pp 68–74https://doi.org/10.1145/3645259.3645271In image classification methods, the quality of the input image plays an important role in improving classification performance. However, sometimes the low resolution and sharpness of remote sensing images can cause various problems in image analysis. ...
- research-articleApril 2024
Highly compressed image representation for classification and content retrieval
Integrated Computer-Aided Engineering (ICAE), Volume 31, Issue 32024, pp 267–284https://doi.org/10.3233/ICA-230729In this paper, we propose a new method of representing images using highly compressed features for classification and image content retrieval – called PCA-ResFeats. They are obtained by fusing high- and low-level features from the outputs of ResNet-50 ...
- research-articleApril 2024
EIGAN: An explicitly and implicitly feature-aligned GAN for degraded image classification
Pattern Recognition Letters (PTRL), Volume 178, Issue CFeb 2024, pp 195–201https://doi.org/10.1016/j.patrec.2024.01.012AbstractThe implementation of classification networks encounters a substantial decline in performance when subjected to degraded images due to factors such as blur, noise, and low resolution. Existing methods focus on addressing a specific kind of ...
Highlights- EIGAN(ours) can guide the model to learn features that are more consistent with high-quality images.
- EIGAN improves the classification accuracy on degraded images without any additional parameters.
- With the deepening of image ...
- research-articleApril 2024
Dynamic semantic structure distillation for low-resolution fine-grained recognition
AbstractLow-resolution images are ubiquitous in real applications such as surveillance and mobile photography. However, existing fine-grained approaches usually suffer catastrophic failures when dealing with low-resolution inputs. This is because their ...
Highlights- Introducing the Dynamic Semantic Structure Distillation (DSSD) framework for enhanced fine-grained image classification in low-resolution images.
- Proposing dynamic semantic structure learning for perceiving semantic relationships, and ...
- research-articleApril 2024
Investigating the effectiveness of data augmentation from similarity and diversity: An empirical study
AbstractData augmentation has emerged as a widely adopted technique for improving the generalization capabilities of deep neural networks. However, evaluating the effectiveness of data augmentation methods solely based on model training is ...
Highlights- We propose quantitative measures to investigate the effectiveness of DA methods.
- Our quantitative measures formulate the similarity and diversity metrics for DA.
- The proposed measures are conducted in feature space, rather than raw ...
- research-articleApril 2024
An interpretable image classification model Combining a fuzzy neural network with a variational autoencoder inspired by the human brain
Information Sciences: an International Journal (ISCI), Volume 661, Issue CMar 2024https://doi.org/10.1016/j.ins.2023.119885AbstractFuzzy neural networks (FNNs) have gained attention for their interpretability and self-learning ability. However, they struggle with interpreting high-dimensional unstructured data and the problem of “rule explosion”. To address this, a model ...
- research-articleApril 2024
Advanced Analysis of Bipolar Disorder Through Computer Vision Technology
Wireless Personal Communications: An International Journal (WPCO), Volume 134, Issue 4Feb 2024, pp 2101–2120https://doi.org/10.1007/s11277-024-10992-wAbstractThis research paper focuses on the analysis of Bipolar Disorder (BD) through the application of structural magnetic resonance imaging (MRI) and a supervised learning framework. The primary goal is to identify specific abnormalities in brain cells ...
- research-articleApril 2024
Dynamic Ensemble Learning with Evolutionary Programming and Swarm Intelligence for Image Classification
Procedia Computer Science (PROCS), Volume 230, Issue C2023, pp 669–678https://doi.org/10.1016/j.procs.2023.12.122AbstractImage classification is a fundamental and pervasive task in the field of computer vision, with profound implications across a wide array of applications. From the autonomous vehicles navigating complex urban environments to the critical diagnosis ...
- research-articleApril 2024
Soft Hybrid Knowledge Distillation against deep neural networks
AbstractTraditional knowledge distillation approaches are typically designed for specific tasks, as they primarily distilling deep features from intermediate layers of a neural network, generally with ingeniously designed knowledge representations, which ...
- research-articleApril 2024
Architecture search of accurate and lightweight CNNs using genetic algorithm
Genetic Programming and Evolvable Machines (KLU-GENP), Volume 25, Issue 1Jun 2024https://doi.org/10.1007/s10710-024-09484-4AbstractConvolutional neural networks (CNNs) are popularly-used in various AI fields, yet the design of CNN architectures heavily depends on domain expertise. Evolutionary neural architecture search (ENAS) methods can search for neural architectures ...
- research-articleMarch 2024
Automated Digitization of Student’s Marks from the Answer-Book Images Using a Lightweight CNN Model
AbstractPreparing student’s digital marksheet using images of student answer-books is a potential application in academic institutions. Segmenting assigned marks automatically from answer-book images is extremely challenging, and it also demands pre-...