A5049170681- Autonomous Vehicle Technology and Safety
- Intelligent Tutoring Systems and Adaptive Learning
- Domain Adaptation and Few-Shot Learning
- Vehicle emissions and performance
- Machine Learning and Algorithms
- Gaussian Processes and Bayesian Inference
- Traffic and Road Safety
- Recommender Systems and Techniques
- Vehicle Dynamics and Control Systems
- Authorship Attribution and Profiling
- Video Surveillance and Tracking Methods
- Computational and Text Analysis Methods
Chengdu Medical College
2025
Beijing Institute of Technology
2023-2024
ABSTRACT Classifying tennis movements from video data presents significant challenges, including overfitting, limited datasets, low accuracy, and difficulty in capturing dynamic, real‐world conditions such as variable lighting, camera angles, complex player movements. Existing approaches lack robustness practicality for real‐time applications, which are crucial sports analysts coaches. To address these this paper proposes an advanced architecture that strategically integrates the...
Online driving style recognition can enhance the customization of human-centric systems, thereby improving comfort, safety, and fuel economy. However, limited performance automotive-grade chips makes it highly challenging to compile run complicated algorithms in real time. To overcome this bottleneck, paper proposes an embedded method for recognizing styles, which is computationally efficient. This approach leverages experts' prior knowledge learning applies electronic control unit (ECU)...
Driving style is usually used to characterize driving behavior for a driver <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">or</i> group of drivers. However, it remains unclear how one individual's shares certain common grounds with other Our insight that sequence responses the weighted mixture latent styles are shareable xmlns:xlink="http://www.w3.org/1999/xlink">within</i> and xmlns:xlink="http://www.w3.org/1999/xlink">between</i>...
Effective driving style analysis is critical to developing human-centered intelligent systems that consider drivers' preferences. However, the approaches and conclusions of most related studies are diverse inconsistent because no unified datasets tagged with styles exist as a reliable benchmark. The absence explicit labels makes verifying different algorithms difficult. This paper provides new benchmark by constructing natural dataset Driving Style (100-DrivingStyle) subjective evaluation...
Driving style recognition plays a vital role in devel-oping human-centered intelligent vehicles that consider drivers' preferences. However, the feature selection of driving is diverse and inconsistent, which varies with scenarios. Therefore, application limited by accuracy rapidity scene algorithm, difficult for low-cost onboard chips. To solve problem, this paper proposes scene-insensitive method recognition. Factor analysis employed to extract common factors scenes from high-dimensional...
Driving style is usually used to characterize driving behavior for a driver or group of drivers. However, it remains unclear how one individual's shares certain common grounds with other Our insight that sequence responses the weighted mixture latent styles are shareable within and between individuals. To this end, paper develops hierarchical model learn relationship styles. We first propose fragment-based approach represent complex sequential behavior, allowing sufficiently representing in...