Research on Dynamic Virtual Fitting Method based on Kinect and Unity3D

The thesis obtains the data with the help of Kinect, combined with the platform of unity3D to research 3D virtual fitting. Using the stroke and a matching algorithm of hidden Markov chain model to establish the corresponding relationship between contour of the human body and the human body model. It is based on multimedia devices to access and the unity of the human body skeleton control method to realize real-time fitting. Combined with Kinect method, 3D virtual character models deformation and skeletal binding methods to achieve a breakthrough in the traditional Unity3D, interact naturally through simple body movements, can be well applied to the virtual fitting system.