Biometric gait verification by horizontal swings in frontal manner towards human-aware environment

Abstract

Over half a century has passed since the first proposal of 'human-computer symbiosis' by Licklider[1]. However people cannot yet receive sufficient and suitable services without their proactive actions because systems cannot recognize dynamic situation in the Real Space. In order to develop a human-aware environment in the name of 'human-computer symbiosis', we target human verification problem and develop gait verification function by movement time series based on body-skeleton information. In this paper, we employ a portable video game device 'Kinect c ', which is used for measuring human body-skeleton information. As the result of gait verification simulation, we show that our developed function works successfully in spite of simple approach and an easy-to-use device.

Soft biometric verification by integrating static and dynamic features based on fuzzy inference

“…Especially the concern with soft biometric traits has been growing [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14]. Soft biometrics is defined as a set of traits providing information about an individual, though these are not able to individually authenticate the subject because they lack distinctiveness and permanence [7].…”

Multi-angle Gait Recognition Based on Skeletal Tracking Data

“…Some recent researches have challenged gait recognition from the front view [16], [17], [18], [19], [20], [21], [22]. Soriano et al claims that front-view gait recognition has an advantage that it is compatible with the ability of humans to recognize their fellow beings from their front-view gaits [16].…”