A5045034384- Multimodal Machine Learning Applications
- Human Pose and Action Recognition
- Advanced Neural Network Applications
- Domain Adaptation and Few-Shot Learning
- Ultrasound Imaging and Elastography
- Computational Drug Discovery Methods
- Medical Image Segmentation Techniques
- Cavitation Phenomena in Pumps
- Advanced Control Systems Optimization
- Metabolomics and Mass Spectrometry Studies
- Advanced Chemical Sensor Technologies
- Aerospace and Aviation Technology
- Lignin and Wood Chemistry
- Human-Automation Interaction and Safety
- Electric Motor Design and Analysis
- Aerospace Engineering and Control Systems
- Cultural Heritage Materials Analysis
- Diet and metabolism studies
- Fungal Biology and Applications
- Computational Fluid Dynamics and Aerodynamics
- Enzyme-mediated dye degradation
- Anomaly Detection Techniques and Applications
- Magnetic Bearings and Levitation Dynamics
- Virtual Reality Applications and Impacts
- Aerospace Engineering and Applications
Chinese University of Hong Kong
2024
California University of Pennsylvania
2024
Beijing University of Technology
2023
Centro Universitário FEI
2023
Huaiyin Normal University
2019
University of Alabama
2001
Abstract Background Alzheimer’s disease (AD) is a chronic neurodegenerative disorder that poses substantial economic burden. The Random forest algorithm effective in predicting AD; however, the key factors influencing AD onset remain unclear. This study aimed to analyze lipoprotein and metabolite using machine-learning methods. It provides new insights for researchers medical personnel understand reference early diagnosis, treatment, prevention of AD. Methods A total 603 participants,...
Attention-based Transformer models have achieved remarkable progress in multi-modal tasks, such as visual question answering. The explainability of attention-based methods has recently attracted wide interest it can explain the inner changes attention tokens by accumulating relevancy across layers. Current simply update equally token before and after processes. However, importance values is usually different during relevance accumulation.In this paper, we propose a weighted strategy, which...
The recent advance in neural rendering has enabled the ability to reconstruct high-quality 4D scenes using networks. Although reconstruction is popular, registration for such representations remains a challenging task, especially dynamic scene surgical planning and simulation. In this paper, we propose novel strategy registration. We first utilize Gaussian Splatting represent capture both static effectively. Then, spatial aware feature aggregation method, Spatially Weight Cluttering (SWC)...
Machine learning algorithms advance the classification of cultural relics in recent studies. A part studies adopts image processing methods as a solution. We believe that blurring limits this approach. On other hand, we notice conducted from perspective chemical composition, which avoids problem images and shows interpretability. In paper, follow direction classifying by composition aim to provide our method with better Hence, propose strategy combined both supervised weakly It visualizes...
Attention-based transformer models have achieved remarkable progress in multi-modal tasks, such as visual question answering. The explainability of attention-based methods has recently attracted wide interest it can explain the inner changes attention tokens by accumulating relevancy across layers. Current simply update equally token before and after processes. However, importance values is usually different during relevance accumulation. In this paper, we propose a weighted strategy, which...
We present Ego-Exo4D, a diverse, large-scale multimodal multiview video dataset and benchmark challenge. Ego-Exo4D centers around simultaneously-captured egocentric exocentric of skilled human activities (e.g., sports, music, dance, bike repair). More than 800 participants from 13 cities worldwide performed these in 131 different natural scene contexts, yielding long-form captures 1 to 42 minutes each 1,422 hours combined. The nature the is unprecedented: accompanied by multichannel audio,...
Three rotor core shaping methods, i.e. asymmetric notching, partial and full shaping, are applied to the design optimization of interior permanent magnet synchronous machines for electric vehicles. The designs also compared with their symmetrical counterparts. It is shown that without sacrificing average torque, all techniques can reduce torque ripples, further output ripple, a as effective fully shaped having increase complexity optimization.