Title :
3D Human Postures Recognition Using Kinect
Author :
Xiao, Zheng ; Mengyin, Fu ; Yi, Yang ; Ningyi, Lv
Author_Institution :
Sch. of Autom., Beijing Inst. of Technol., Beijing, China
Abstract :
In many application cases, 2D human postures display haven´t been able to meet people´s requirements which is failure to show human motions comprehensive, image and vivid. However, 3D human Postures display could restore and show human motions well, which is convenient for people to observe and learn human motions. This paper presents a method to recognize 3D human postures by using Microsoft Kinect sensor. Kinect is used as a capturing device. Capturing 3D human features mainly uses depth images obtained from Kinect sensor. Each pixel of depth images contains three-dimensional coordinate information of camera´s scenes. Finally, the captured 3D human postures can be displayed by employing a human skeletal joints model and using a LED cube.
Keywords :
cameras; gait analysis; image motion analysis; image recognition; interactive systems; light emitting diodes; 2D human posture display; 3D human features; 3D human posture display; 3D human posture recognition; LED cube; Microsoft Kinect sensor; camera scenes; depth image pixels; human motions; human skeletal joint model; three-dimensional coordinate information; Cameras; Humans; Joints; Light emitting diodes; Robot sensing systems; Solid modeling; Vegetation; 3D human postures; Kinect; LED cube; depth images; human motions; human skeletal joints model;
Conference_Titel :
Intelligent Human-Machine Systems and Cybernetics (IHMSC), 2012 4th International Conference on
Conference_Location :
Nanchang, Jiangxi
Print_ISBN :
978-1-4673-1902-7
DOI :
10.1109/IHMSC.2012.92
