基于Kinect三维行动重建应用于传统武术教学训练研究

Research on Application of Kinect Three-dimensional ActionReconstruction to Traditional Martial Arts Teaching and Training

在传统武术教学训练中,运动技能学习可以通过视频观察模仿学习与练习在目标运动中的行为来促进运动技能学习。当学习者使用视频图像观察动作时,视点改变函数是可取的。然而,一般来说,观察的视点改变是不可能的,因为大多数视频是使用单一摄像机从固定点拍摄的。本研究的目的是开发一种方法,从单个图像中生成模拟仿真人体运动的3D人体模型,并能够从任何角度观察虚拟练习者众所周知,的行为。作为第一步,本研究以传统武术训练为重点,开发了一种半自动的传统武术拳击视频图像的三维重建方法。

In traditional martial arts teaching and training,sports skills learning can promote motor skills learning by making video observation and imitating the behavior in target sports.The viewpoint change function is applicable when learners observe actions using video images.However,in general,the observed viewpoint changes are impossible because most videos were taken from a fixed point using a single camera.The aim of this study is to develop a method to generate 3D mannequins simulating human motion from individual images and to be able to observe the behavior of virtual practitioners from any perspective.As a first step,this study developed a semi-automatic 3-dimensional reconstruction of video images of traditional boxing.