A fast garment fitting algorithm using skeleton‐based error metric

Abstract We present a fast and automatic method to fit a given 3D garment onto a human model with various shapes and poses, without using a reference human model. Our approach uses a novel skeleton‐based error metric to find the pose that best fits the input garment. Specifically, we first generate the skeleton of the given human model and its corresponding skinning weights. Then, we iteratively rotate each bone to find its best position to fit the garment. After that, we rig the surface of the human model according to the transformations of the skeleton. Potential penetrations are resolved using collision handling and physically based simulation. Finally, we restore the human model back to the original pose in order to obtain the desired fitting result. Our experiment results show that besides its efficiency and automation, our method is about two orders of magnitudes faster than existing approaches, and it can handle various garments, including jacket, trousers, skirt, a suit of clothing, and even multilayered clothing.