{"title":"A new approach for tracking human body movements by kinect sensor","authors":"Adjeisah Michael, Zhao Chen, Guohua Liu, Yang Yi","doi":"10.1049/CP.2017.0117","DOIUrl":null,"url":null,"abstract":"With the invention of the Microsoft Kinect sensor, high-resolution depth and visual sensing has become available for prevalent use in the smart home and medical rehabilitation. In this paper, we introduce an innovative approach for effective body movement tracking using Kinect Xbox 360 with a limited tracking system. A relatively scaled hand cursor mechanism is used for our system of interaction. Instead of tracking the whole body of the participant in the Kinect Depth space which produces 20 joints, we limit the tracking to only 2 joint (left and right hand) for the same action. Further, we have engaged Extended Kalman Filter to improve skeleton joint estimation which smooths the joint coordinates, placing the Z axis in a high level of calibration in order to make it work with X and Y coordinates simultaneously with a relatively high accuracy.","PeriodicalId":424212,"journal":{"name":"4th International Conference on Smart and Sustainable City (ICSSC 2017)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"4th International Conference on Smart and Sustainable City (ICSSC 2017)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1049/CP.2017.0117","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
With the invention of the Microsoft Kinect sensor, high-resolution depth and visual sensing has become available for prevalent use in the smart home and medical rehabilitation. In this paper, we introduce an innovative approach for effective body movement tracking using Kinect Xbox 360 with a limited tracking system. A relatively scaled hand cursor mechanism is used for our system of interaction. Instead of tracking the whole body of the participant in the Kinect Depth space which produces 20 joints, we limit the tracking to only 2 joint (left and right hand) for the same action. Further, we have engaged Extended Kalman Filter to improve skeleton joint estimation which smooths the joint coordinates, placing the Z axis in a high level of calibration in order to make it work with X and Y coordinates simultaneously with a relatively high accuracy.
随着微软Kinect传感器的发明,高分辨率深度和视觉传感已经在智能家居和医疗康复中得到广泛应用。在本文中,我们介绍了一种利用Kinect Xbox 360有限跟踪系统进行有效身体运动跟踪的创新方法。我们的交互系统使用了一个相对缩放的手光标机制。比起在Kinect深度空间中追踪参与者的整个身体(游戏邦注:这会产生20个关节),我们将同一动作的追踪限制为只有2个关节(左手和右手)。此外,我们还使用扩展卡尔曼滤波器来改进骨架关节估计,使关节坐标平滑,将Z轴置于高水平的校准中,以使其与X和Y坐标同时工作,具有相对较高的精度。