Studies Using My Datasets
This curated list links to research that reports using data from MultiSenseBadminton, Engagnition, or TimelyTale. It is not a citation count or an exhaustive list of citing papers. Work I co-authored is listed separately from other publications.
MultiSenseBadminton
Multimodal wearable and motion data for analyzing badminton strokes and skill levels.
Source paper · Open dataset · Google Scholar citations
Other publications using the dataset
- Physical Fitness and Tactical Optimization of Badminton Based on Multimodal Fusion and Swarm Intelligence Algorithm
Uses the dataset's inertial streams for cross-subject badminton stroke recognition and offline tactical analysis. Usage evidence - A Deep Learning Framework with Kinetic Chain Graphs for Sports Striking Motion Evaluation
Evaluates a kinetic-chain graph model on the dataset's forehand-clear and backhand-drive strokes. - ChainBMD: A Deep Learning Framework for Stroke Evaluation and Guidance Using Kinetic Chain Graphs and Temporal Analysis
Combines foot pressure, EMG, and joint-angle streams to evaluate badminton strokes and generate guidance. Usage evidence - Gaze Behaviors During Forehand Clear and Backhand Driving in Badminton: A Comparison Between Beginner, Intermediate and Expert Players
Reanalyzes the released eye-tracking and annotation data across stroke types and skill levels.
Co-authored follow-up work
- TS2Vec-based Time-Series Representation Learning for Badminton Swing Skill Level Assessment
Uses 3D skeleton sequences and expert skill labels for swing-level assessment. - Counterfactual Explanation-Based Badminton Motion Guidance Generation Using Wearable Sensors
Uses joint positions and stroke-quality labels to generate motion guidance. - Badminton Swing Posture Analysis and Visualization Using Deep Learning-Based Encoder-Decoder Architecture
Uses joint trajectories and swing annotations to visualize posture differences.
The IJCNN and ChainBMD papers are related outputs from the same research group.
Engagnition
Physiological, movement, performance, and expert-annotated engagement data from 57 children with autism spectrum disorder.
Source paper · Open dataset · Google Scholar citations
Other publications using the dataset
- Transfer Learning from Datasets with Unshared Features for Detecting Autism Spectrum Disorder Using AutDB
Uses Engagnition wristband signals and annotations in a cross-dataset movement-analysis benchmark. Usage evidence
TimelyTale
Multimodal driving and passenger-state data documenting when passengers request explanations in highly automated vehicles.
Other publications using the dataset
- What Amplifies Explanation Needs in Autonomous Vehicles?: Feature Interaction Between Maneuver Dynamics and Passenger Physiological Responses
Reanalyzes TimelyTale vehicle-dynamics and passenger-physiology features with SHAP interaction analysis to examine when explanation needs increase.