A5011321978- Infrastructure Maintenance and Monitoring
- Multimodal Machine Learning Applications
- Subtitles and Audiovisual Media
- Traffic Prediction and Management Techniques
- Occupational Health and Safety Research
- Non-Destructive Testing Techniques
- Railway Systems and Energy Efficiency
- Speech and dialogue systems
- Human Pose and Action Recognition
- BIM and Construction Integration
- Concrete Corrosion and Durability
- Railway Engineering and Dynamics
University of Michigan
2024
Rutgers Sexual and Reproductive Health and Rights
2024
Yonsei University
2020-2022
Abstract Zero‐shot learning, applied with vision‐language pretrained (VLP) models, is expected to be an alternative existing deep learning models for defect detection, under insufficient dataset. However, VLP including contrastive language‐image pretraining (CLIP), showed fluctuated performance on prompts (inputs), resulting in research prompt engineering—optimization of improving performance. Therefore, this study aims identify the features a that can yield best classifying and detecting...
Construction tasks impose high ergonomic risks, making it crucial to observe problems (e.g., actions and postures associated with risks) provide solutions. As manually identifying solutions is time-consuming subjective, there has been extensive development toward automation through computer vision-based or sensor-based applications. Nevertheless, most existing studies have focused on assessing leaving of recognizing generating ergonomists. However, ergonomists are scarce in construction....
Work-related musculoskeletal disorders (WMSDs) are a primary cause of non-fatal injuries in diverse industries. Traditional manual ergonomic problem and solution identification for reducing WMSDs is time-consuming limited by expert availability. Image captioning—interpreting images workers their workplaces capturing interactions therein—is one potential alternative. Yet, due to the absence knowledge, conventional image captioning models generating accurate captions problems solutions....
This study reviewed the status quo of research on machine learning (ML), including deep learning, in rail industry. conducted a scientometric analysis and critical review 640 papers selected from 12,675 web-crawled papers. The extensive complex networks topics, researchers, countries were analyzed using Louvain method, co-occurrence keyword analysis, degree centrality other network methods. results indicate that majority studies ML industry focused maintenance activities traffic management,...