Image-based dress up system

In this work, we present an image based virtual dress up system according to user input model and garment image. At 'Registration' step, we asked the user manually setting the skeleton structure and matte out the alphamap from the image. Next step, our method automatically deforms the garment image corresponding to model's body. For the boundary fitting, ourmethod uniformly sampled contour points and solves the optimization function. To enhance the more realistic scene, we reconstruct the 2D mesh to the 3D mesh according to a human's standard body shape. For the lighting effect we estimate the light position by using luminance value with the detected face region. Previous 3D scanner based virtual dress up system has expensive cost and under locational limitation issues, but our system integrates various image processing techniques and introduced an easy-to-use system for the general users. We present that our system produces a visually plausible and well-fitted virtual dress up results in a practical and usable way.