With the expansion of animation industry, there has been a recurring issue of animation plagiarism. Currently, the technology for copyright protection and plagiarism detection in animation is limited. This paper proposes a framework of plagiarism ...

• A novel framework called R-Net is proposed for salient object detection.
• Long-range dependency solves space context separation within feature map.
• Recursive decoder generates preliminary salient map by full-scale feature fusion.

Fully convolutional neural (FCN) networks based on encoder-decoder structures are adopted for salient object detection (SOD) and achieve state-of-the-art performance. However, most of the previous works aggregate multilevel features layer by ...

Few-shot segmentation (FSS) has made remarkable success through prototypical learning. However, owing to the scarcity of support data, FSS methods continue to suffer from huge intra-class and inter-class gaps. In this paper, we introduce a ...

• This paper is extented by our conference paper in IJ- CAI’23 with 50% new content.
• We propose FGNet for FSS to fill intra-class and inter- class gaps.
• A novel Feature Transformation Module is proposed for feature enhancement.
• ...

Multi-instance learning is a classic algorithm within the weakly supervised learning framework. Most previous approaches to multi-instance learning have typically assumed that every bag comes with its own label. However, in real-world ...

• MIKCE tackles problem as a set of convex sub-problems for more efficient solving.
• MIKCE is independent of the loss function, square loss and hinge loss are adopted.
• Unlike previous approaches, MIKCE does not rely on the positive ...

In personalized predictive medicine, accurately modeling a patient’s illness and care processes is essential, given their inherent long-term temporal dependencies. However, electronic medical records contain episodic and irregularly timed data ...

• Enhancing patient illness prediction via frequency and recent observation integration.
• Effective handling of irregular timing through time parameterization.
• Adaptive pooling integration to address outliers in patient EHR data.
• ...

The complex and nonlinear nature of brain tumor morphology and texture pose significant challenges for representative feature extraction from magnetic resonance imaging (MRI) images, ultimately leading to inaccurate diagnoses of brain tumors. To ...

The burgeoning popularity of cloud gaming makes it critical for efficient video compression to relieve the growing bandwidth pressure. While existing neural video coding approaches have demonstrated strong compression potential on natural videos, ...

• Efficient neural video compression for cloud gaming videos utilizing camera motion.
• An aligned frame generation network to improve the accuracy of motion estimation.
• An enhanced latent prior for the entropy model to predict a ...

• Epistemic uncertainty reveals the effects of dataset imbalance in machine learning.
• Uncertainty quantification is an effective tool for guiding a resampling strategy.
• Improving the quality of uncertainty quantification metrics ...

Several factors are known to determine the quality of machine learning models, one of which is the dataset quality. One problem related to the quality of a dataset is the imbalance issue. An imbalanced dataset contains significantly more data ...

Distributed control systems (DCS) are essential to operate complex industrial processes. A major part of a DCS is the alarm system, which helps plant operators to keep the processes stable and safe. Alarms are defined as threshold values on ...

• Multimodal Transformers can successfully predict alarms in industrial settings.
• Transformers can predict future alarms based on recent events and signal data.
• The proposed hybrid fusion performs on par with or better than early or ...

Deep neural networks (DNNs) present significant potential as surrogate models to substitute for costly simulations. Attaining high accuracy in DNNs is crucial for effective surrogate modeling. However, this goal necessitates the availability of ...

Belief rules are an extension of fuzzy rules that consider belief degrees. They are widely applied due to their traceability, interpretability, and flexibility. However, current belief rules only consider residual support and global ignorance ...

• Introduces belief rules with partial ignorance to enhance uncertainty modeling.
• Proposes a learning algorithm balancing accuracy and complexity.
• Enhances inference method using Dempster’s rule and Shafer’s discounting.
• ...

Bitcoin’s volatile nature has made its price prediction a sought-after mathematical model in the FinTech industry. Existing studies, however, need to look into the critical aspect of time-lagged sentiment in Bitcoin price forecasting. This ...

In recent years, information consumption among the population has increased, especially through the widespread use of digital images and videos. Consequently, the demand for efficient compression methods has escalated, aiming to facilitate both ...

Parkinson’s disease (PD) is a prevalent chronic neurodegenerative disorder characterized by both motor and non-motor symptoms. The significant heterogeneity among PD patients poses a major challenge for treatment interventions. Current clinical ...

Due to the complex backgrounds and diverse attributes of salient objects in optical remote sensing images (RSIs), existing methods heavily rely on pixel-level annotations, with sparse annotations receiving limited attention. Furthermore, the ...

• MIFA-Net: Novel multi-source fusion network for weakly supervised salient object detection.
• Effective Fusion: Integrates sparse annotations, enhancing feature representation for high-quality saliency maps.
• Innovative Modules: BDB, ...

Image and associated features detect behavioural anomalies in crowd scenes through matching. Existing methods use continuous data inputs to track crowd behaviour. As a result, the system mistakenly assumes normal behaviour during data gaps, ...

• A novel data fusion technique model based on deep learning was suggested.
• Various models were utilized in the forecasting of daily DO levels in aquatic ecosystems.
• MI and RFE methods were applied as feature selection strategies.

Accurate monitoring of dissolved oxygen (DO) levels is critical for stakeholders to effectively safeguard water resources and aquatic ecosystem health. This research presents an innovative data fusion framework based on Bayesian model averaging (...

Self-supervised learning in the federated learning paradigm has been gaining a lot of interest both in industry and research due to the collaborative learning capability on unlabeled yet isolated data. However, self-supervised based federated ...

In person re-identification(Re-ID) tasks, using only convolutional neural network(CNNs) makes it challenging to fully exploit the global key information in person images, while using only the Transformer structure fails to adequately capture the ...

• A convolutional neural network (MCF) incorporating an attention mechanism.
• Transformer feature aggregator structures (SDT) based on single and dual channels .
• Use SDT to aggregate features between MCF layers to form an ADT model.

This article proposes the concept of two-dimensional picture fuzzy sets (TDPFS). This new framework extends the original picture fuzzy sets (PFS) by incorporating a confidence level to evaluate expert opinions using probability. This innovation ...