VSGD: a Bi-modal Dataset for Gait Analysis

Gait refers to the displacement of the center of gravity during motion. Gait analysis has been a focus of much research in recent years. However, the literature on gait analysis either deals with visual gait data or with inertial gait data collected using wearable sensors. In this work, we propose a new dataset collected on our campus of 45 subjects (32 males and 13 females) of ages from 18 to 23 walking a straight path while wearing 4 inertial measurement units and being filmed using two smartphones fixed at two different directions. Both the visual and inertial data are recorded at the same time, and an asynchronous signal was performed by every subject to be able to align both modalities together. We validate our data on two classification models for gender and person recognition and show that the two models perform well on our data.