A5101901155- Anomaly Detection Techniques and Applications
- Digital Games and Media
- Social Robot Interaction and HRI
- Emotion and Mood Recognition
- Face recognition and analysis
- Computer Graphics and Visualization Techniques
- Image Processing and 3D Reconstruction
- Blockchain Technology Applications and Security
- Advanced Algorithms and Applications
- Multimedia Communication and Technology
- Geological Modeling and Analysis
- Dynamics and Control of Mechanical Systems
- Product Development and Customization
- Hydraulic and Pneumatic Systems
- Network Security and Intrusion Detection
- Metaheuristic Optimization Algorithms Research
- Fault Detection and Control Systems
- Digital Mental Health Interventions
- Hydrocarbon exploration and reservoir analysis
- Manufacturing Process and Optimization
- Advanced machining processes and optimization
- Privacy, Security, and Data Protection
- AI in Service Interactions
- Imbalanced Data Classification Techniques
- Design Education and Practice
Zhejiang University
2021-2024
South China University of Technology
1987-2023
Ningxia University
2012-2023
Zhejiang University of Science and Technology
2022
Xiaomi (China)
1997
Hengyang Academy of Agricultural Sciences
1991
AI drawing tools, which rely on high-quality prompts to guide and refine the creative process, have potential enhance efficiency quality during conceptual design stage. However, addressing inherent accuracy issues of is crucial. This study explores usability integrating with prompt optimisation tools introduces a GPT-based tool for validation. Experiment results indicate that, divergent thinking stage concept design, both junior senior designers using this reduced their average time...
We examined the impact of firm-generated content on firm-related user-generated content. Researches have proven that impacts firm revenue. Therefore, it is necessary to determine what factors stimulate creation Creating videos related companies (e.g. E-sports) Chinese online video platforms often includes use as material. It may suggest play an essential role in influencing Internet users create collected a unique dataset using Python web crawler and used LSDV model for empirical analysis,...
Empowering retail service robots with empathy is one of the current research hotspots in field artificial intelligence. Identifying consumer emotions, understanding changes shopping interests, and developing appropriate sales strategies a challenging task for robots. We investigate feasibility using computer vision methods empowering by examining correlation between emotion levels interest. To this end, we construct first video database sentiment business context propose deep learning method...
Faced with the impact of coronavirus disease (COVID-19) pandemic, governments must protect well-being population. Aside from considerations, such as keeping virus spreading and treating patients, government should also be concerned about mental health its citizens during epidemic. This study aimed to help users who develop depression due COVID-19 on social media, reduce cost counselling, need for visit hospital counselling. investigated opportunities blockchain technology provide...
Interactive videos have been applied to various areas due their engagement potential and efficiency improvement of information communication. However, creating interactive can be challenging because a lack novice-oriented guidance in current platforms, the logic-building process when authoring videos. To address these challenges, we obtained insights from four creativity support tool designers, proposed series hierarchical video structures based on existing narrative structures, presented...
How to make robots more human-like has been a hot topic of interest for researchers in the field robot interaction. Empowering retail services with empathy read minds is still major challenge. To address this problem, we proposed deep learning-based method programs infer consumers' shopping levels, which confirmed effectiveness met or exceeded that human judgement. The experimental results demonstrated achieved an accuracy 65% predicting consumer compared 46% levels manual assessment (n =...
A common study area in anomaly identification is industrial images detection based on texture background. The interference of and the minuteness anomalies are main reasons why many existing models fail to detect anomalies. We propose a strategy for that combines dictionary learning normalizing flow aforementioned questions. two-stage approach already use enhanced by our method. In order improve baseline method, this research adds representation deep learning. Improved algorithms have...
Evolutionary Programming (EP) is a kind of stochastic optimization algorithm. The goal EP to achieve intelligent behavior through simulated evolution. algorithms are based on an arbitrarily initialized population search points which evolves towards better and regions in the space by means randomized process mutation selection. To avoid premature convergence balancing ability exploration exploitation has become one important aspects EP's study. We describe classic evolutionary programming...
A common study area in anomaly identification is industrial images detection based on texture background. The interference of and the minuteness anomalies are main reasons why many existing models fail to detect anomalies. We propose a strategy for that combines dictionary learning normalizing flow aforementioned questions. two-stage approach already use enhanced by our method. In order improve baseline method, this research add representation deep learning. Improved algorithms have exceeded...