| With in-depth application of computer technologies such as digital imageprocessing, virtual reality in the medical field, digital medical as a cross subject ofdigital art and medicine has achieved rapid development in recent years. As animportant part in the field of medicine, medical teaching also gradually changed to thedirection of digital and virtualization.In medical teaching human anatomy is an important basic course and a requiredsubject in the study of traditional Chinese medicine and western medicine. Existingforms include autopsy teaching in practice, using the map and specimens of humanorgans as display, using a digital virtual human teaching and so on. But traditionalautopsy method faced with the problem of insufficient body resources in recent years,and the chemical composition of its preservative can harm the students’ health underthe condition of suction long time. In addition, some students use map and specimensshow, but this method lacks depth and interactivity, difficult to stimulate students’interest in learning; Most schools abroad and a few domestic schools have begun touse digital virtual human teaching, but most use traditional devices such as mouse,keyboard and touch screen in teaching, also lack interactivity and enjoyment.In order to solve the above problems, this paper has developed a virtual humanbody anatomy teaching system which uses the three-dimensional interaction softwareUnity3D and Kinect gesture recognition technology. Using this system can realizemany people view, learning and research, there is no shortcomings such as lack ofresources, chemical pollution and so on, it’s interactive, with mobile convenience,users can scale and rotate the body models arbitrarily, and can watch themindividually, enhance the convenience and fun of students’ learning. This system useUnity3D software as the main development platform, through the study and analysisof3D scene rendering system and script control system in Unity3D, realized themodel importing and the connection with Kinect, then obtain the data from Kinect asinput to control the display of the human body model. Meanwhile, this paper analysesthe current commonly used methods to access gesture interaction with Kinect,including using OpenNI, OpenNI2and Microsoft Kinect for Windows SDK, thenmakes the encapsulate and extension of Kinect for Windows SDK in Unity3D on thebasis, defined some commonly used gesture interfaces, realized the real-timeidentification of the gestures such as grip, release, slide to the left or right, handsmove open or closed, the corresponding gesture is used to control the capture, free,move left or right, zoom in or out operation of the model. After tested, system realizedthrough the above methods showed high practicability, interactivity and enjoyment. |