Metaverse Gait Authentication Dataset (MGAD)

<b>1. Dataset Overview</b><br>The <b>Metaverse Gait Authentication Dataset (MGAD)</b> is a large-scale dataset for gait-based biometric authentication in virtual environments. It consists of gait data from <b>5,000 simulated users</b>, generated using <b>Unity 3D</b> and processed using <b>OpenPose and MediaPipe</b>. This dataset is ideal for researchers working on biometric authentication, gait analysis, and AI-driven identity verification systems.<b>2. Data Structure &amp; Format</b><b>File Format:</b> CSV<b>Number of Samples:</b> 5,000 users<b>Number of Features:</b> 16 gait-based features<b>Columns:</b> Each row represents a user with corresponding gait feature values<b>Size:</b> Approximately (mention size in MB/GB after upload)<b>3. Feature Descriptions</b>The dataset includes 16 extracted gait features:<b>Stride Length (m)</b>: Average distance covered in one gait cycle.<b>Step Frequency (steps/min)</b>: Number of steps taken per minute.<b>Stance Phase Duration (s)</b>: Stance phase in a gait cycle.<b>Swing Phase Duration (s)</b>: Duration of the swing phase in a gait cycle.<b>Double Support Phase Duration (s)</b>: Time both feet are in contact with the ground.<b>Step Length (m)</b>: Distance between consecutive foot placements.<b>Cadence Variability (%)</b>: Variability in step rate.<b>Hip Joint Angle (°)</b>: Maximum angle variation in the hip joint.<b>Knee Joint Angle (°)</b>: Maximum flexion-extension knee angle.<b>Ankle Joint Angle (°)</b>: Angle variation at the ankle joint.<b>Avg. Vertical GRF (N)</b>: Average vertical ground reaction force.<b>Avg. Anterior-Posterior GRF (N)</b>: Ground reaction force in the forward-backward direction.<b>Avg. Medial-Lateral GRF (N)</b>: Ground reaction force in the side-to-side direction.<b>Avg. COP Excursion (mm)</b>: Center of pressure movement during stance phase.<b>Foot Clearance during Swing Phase (mm)</b>: Minimum height of the foot during the swing phase.<b>Gait Symmetry Index (%)</b>: Measure of symmetry between left and right gait cycles.<b>4. How to Use the Dataset</b>Load the dataset in Python using Pandas:<br>Use the features for machine learning models in biometric authentication.Apply preprocessing techniques like normalization and feature scaling.Train and evaluate deep learning or ensemble models for gait recognition.<b>5. Citation &amp; License</b>If you use this dataset, please cite it as follows:<b>Sandeep Ravikanti, "Metaverse Gait Authentication Dataset (MGAD)," IEEE DataPort, 2025. DOI: https://dx.doi.org/10.21227/rvh5-8842</b><b>6. Contact Information</b>For inquiries or collaborations, please contact: <b>[e-mail]</b>