A5114929938- Landslides and related hazards
- Anomaly Detection Techniques and Applications
- Human Pose and Action Recognition
- Human Motion and Animation
- Tree Root and Stability Studies
- Neural Networks and Applications
- Geotechnical Engineering and Analysis
- 3D Shape Modeling and Analysis
- Advanced Memory and Neural Computing
- Occupational Health and Safety Research
- Video Analysis and Summarization
- Seismology and Earthquake Studies
- Magnetic properties of thin films
- BIM and Construction Integration
- Flood Risk Assessment and Management
- Infrastructure Maintenance and Monitoring
Peking University
2024
Central South University
2022-2024
Onsite systematic monitoring benefits hazard prevention immensely. Hazard identification is usually limited due to the semantic gap. Previous studies that integrate computer vision and ontology can address gap detect onsite hazards. However, extracting encoding regulatory documents in a computer-processable format often requires manual work which costly time-consuming. A novel universally applicable framework proposed integrates vision, ontology, natural language processing improve safety...
Recent advancements in deep convolutional neural networks (DCNNs) have significantly improved landslides identification using remote sensing images. Pixel-wise semantic segmentation (PSS) and object-oriented detection (OOD) are two dominant approaches, wherein PSS better as providing detailed delineation of landslide shapes. However, limited by the difficulty labelling training data low speed compared to OOD. In this paper, we propose an efficient DCNN-based method, so-called...
The improvement of landslide susceptibility assessment is a long-standing problem in hazard mitigation work, wherein previous studies have proposed various training models. However, the ratio positive to negative samples and selection non-landslide been shown significantly influence results. These research directions traditionally focal points, while datasets are often overlooked, serving merely as auxiliary tools support validation process. Hence, this study proposes an approach enhance...
Enabling field-free current-induced switching of perpendicular magnetization is essential for advancing spin-orbit-torque magnetic random access memory technology. Our research on the Pt/Co/Ru/RuO2(101) system has successfully demonstrated through current injection along RuO2[010] axis. We discovered that exhibits a tilted easy axis, inclined from out-of-plane towards RuO2[-101] direction. The application to this axis generates substantial effective field, which facilitates switching....
Synthesizing human motions in 3D environments, particularly those with complex activities such as locomotion, hand-reaching, and Human-Object Interaction (HOI), presents substantial demands for user-defined waypoints stage transitions. These requirements pose challenges current models, leading to a notable gap automating the animation of characters from simple inputs. This paper addresses this challenge by introducing comprehensive framework synthesizing multi-stage scene-aware interaction...