Abstract—Gait has been considered as a promising and unique
2 biometric for person identification. Traditionally, gait data are
3 collected using either color sensors, such as a CCD camera, depth
4 sensors, such as a Microsoft Kinect, or inertial sensors, such as
5 an accelerometer. However, a single type of sensors may only
6 capture part of the dynamic gait features and make the gait
7 recognition sensitive to complex covariate conditions, leading to
8 fragile gait-based person identification systems. In this paper,
9 we propose to combine all three types of sensors for gait data
10 collection and gait recognition, which can be used for important
11 identification applications, such as identity recognition to access
12 a restricted building
2 biometric for person identification. Traditionally, gait data are
3 collected using either color sensors, such as a CCD camera, depth
4 sensors, such as a Microsoft Kinect, or inertial sensors, such as
5 an accelerometer. However, a single type of sensors may only
6 capture part of the dynamic gait features and make the gait
7 recognition sensitive to complex covariate conditions, leading to
8 fragile gait-based person identification systems. In this paper,
9 we propose to combine all three types of sensors for gait data
10 collection and gait recognition, which can be used for important
11 identification applications, such as identity recognition to access
12 a restricted building